content large_stringlengths 3 20.5k | url large_stringlengths 53 192 β | branch large_stringclasses 4
values | source large_stringclasses 51
values | embeddings listlengths 384 384 | score float64 -0.21 0.65 |
|---|---|---|---|---|---|
platforms metadata: hermes: tags: [python, automation] category: devops fallback\_for\_toolsets: [web] # Optional β conditional activation (see below) requires\_toolsets: [terminal] # Optional β conditional activation (see below) config: # Optional β config.yaml settings - key: my.setting description: "What this contro... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
0.007628849241882563,
-0.04378772899508476,
0.03412390872836113,
-0.02355285920202732,
0.031300969421863556,
-0.08793241530656815,
0.04993593320250511,
0.01254906877875328,
-0.0929495319724083,
0.015271774493157864,
0.048964470624923706,
-0.07422854006290436,
0.07114947587251663,
0.0302884... | 0.080752 |
toolsets are available. Shown when they're missing. | | `fallback\_for\_tools` | Same, but checks individual tools instead of toolsets. | | `requires\_toolsets` | Skill is \*\*hidden\*\* when the listed toolsets are unavailable. Shown when they're present. | | `requires\_tools` | Same, but checks individual tools. | \*... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
-0.07299286872148514,
-0.022450275719165802,
-0.08150513470172882,
0.08916575461626053,
0.03484043851494789,
0.006059055216610432,
-0.0004577816289383918,
-0.07163431495428085,
-0.038368768990039825,
-0.04677252098917961,
0.027248715981841087,
-0.07298758625984192,
0.025639917701482773,
0.... | -0.026913 |
files in that directory. Use filesystem permissions or a separate profile/toolset setup if shared external skills must stay read-only. - \*\*Local precedence\*\*: If the same skill name exists in both the local dir and an external dir, the local version wins. - \*\*Full integration\*\*: External skills appear in the sy... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
-0.029432740062475204,
-0.02068832702934742,
-0.03754891827702522,
-0.03478296101093292,
-0.005182732827961445,
-0.08670690655708313,
-0.013006197288632393,
0.0045841047540307045,
-0.028912508860230446,
-0.003874534508213401,
0.03803734481334686,
-0.0022550043649971485,
0.048589032143354416,... | 0.072061 |
loads the skills that do resolve, and the agent gets a note listing what was skipped. - \*\*Bundles work in every surface\*\* β interactive CLI, TUI, dashboard chat, and every gateway platform (Telegram, Discord, Slack, β¦) β because dispatch is centralized in the same place as individual skill commands. - \*\*Bundles d... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
-0.07902567833662033,
-0.03927918151021004,
0.04486633837223053,
0.007950331084430218,
-0.016848035156726837,
-0.051639266312122345,
0.08345799148082733,
0.02839682810008526,
-0.07018034160137177,
0.02073679305613041,
0.03954219073057175,
-0.0292567890137434,
0.05073224753141403,
0.0699022... | 0.091496 |
# Un-stick a bundled skill from "user-modified" (see below) hermes skills reset google-workspace --restore # Also restore the bundled version, deleting your local edits hermes skills publish skills/my-skill --to github --repo owner/repo hermes skills snapshot export setup.json # Export skill config hermes skills tap ad... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
-0.10361454635858536,
-0.029333069920539856,
0.00809724535793066,
0.00359245203435421,
0.018809735774993896,
-0.023011205717921257,
-0.012687234207987785,
-0.02288627438247204,
-0.07551708817481995,
0.023717476055026054,
0.03702902793884277,
0.021947134286165237,
0.05283325910568237,
0.019... | 0.006481 |
publish Claude-compatible plugin/marketplace manifests. Known integrated sources include: - [anthropics/skills](https://github.com/anthropics/skills) - [aiskillstore/marketplace](https://github.com/aiskillstore/marketplace) Hermes source id: `claude-marketplace` #### 7. LobeHub (`lobehub`) Hermes can search and convert... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
-0.08086712658405304,
-0.018544761463999748,
-0.10807554423809052,
-0.0367598794400692,
-0.05480668693780899,
-0.01274002157151699,
-0.012542434968054295,
-0.0029849144630134106,
-0.05921793356537819,
0.026584917679429054,
0.03309403359889984,
-0.025046709924936295,
0.029184965416789055,
0... | 0.140822 |
| | `official` | `optional-skills/` in the repo | Built-in trust, no third-party warning | | `trusted` | Trusted registries/repos such as `openai/skills`, `anthropics/skills`, `huggingface/skills`, `NVIDIA/skills` | More permissive policy than community sources | | `community` | Everything else (`skills.sh`, well-known... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
-0.07402512431144714,
-0.08760640770196915,
-0.07251685112714767,
0.04939012601971626,
0.09664212912321091,
-0.044240616261959076,
-0.02223336510360241,
-0.01697683334350586,
-0.07826682925224304,
0.023280560970306396,
0.03295191004872322,
-0.005041212309151888,
0.0561596043407917,
0.03331... | 0.123192 |
individual skills directly (without adding a tap) Users can also install a single skill from any public GitHub repo without adding the whole repo as a tap: ```bash hermes skills install owner/repo/skills/my-workflow ``` Useful when you want to share one skill without asking the user to subscribe to your whole registry.... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/skills.md | main | hermes-agent | [
-0.04860610514879227,
-0.09612318128347397,
-0.06937776505947113,
-0.015706706792116165,
-0.03946499526500702,
-0.06624051928520203,
0.03302500769495964,
-0.051966894418001175,
-0.04426218196749687,
0.009589769877493382,
0.03727862611413002,
-0.002876679878681898,
0.06863201409578323,
0.06... | 0.032427 |
# Plugins Hermes has a plugin system for adding custom tools, hooks, and integrations without modifying core code. If you want to create a custom tool for yourself, your team, or one project, this is usually the right path. The developer guide's [Adding Tools](/developer-guide/adding-tools) page is for built-in Hermes ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/plugins.md | main | hermes-agent | [
-0.08689486980438232,
-0.03810069337487221,
-0.045744989067316055,
-0.004477895330637693,
-0.06992629170417786,
-0.060284338891506195,
-0.047721583396196365,
0.03108391724526882,
-0.03153041750192642,
0.065486840903759,
0.019332293421030045,
-0.034546028822660446,
0.03012620285153389,
-0.0... | 0.071594 |
| `ctx.llm.complete(...)` / `ctx.llm.complete\_structured(...)` β borrow the user's active model + auth for a one-shot completion with optional JSON schema validation. See [Plugin LLM Access](/developer-guide/plugin-llm-access) | | Register an inference backend (LLM provider) | `register\_provider(ProviderProfile(...))... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/plugins.md | main | hermes-agent | [
-0.0977352112531662,
-0.03006979450583458,
-0.0063277678564190865,
-0.003057660534977913,
0.027580803260207176,
-0.021806582808494568,
-0.014064068906009197,
0.054829180240631104,
-0.07124868780374527,
0.005986763164401054,
0.07092202454805374,
-0.04168042913079262,
0.05597110465168953,
0.... | 0.032787 |
by `memory.provider` in `config.yaml`. | | \*\*Context engines\*\* (`plugins/context\_engine/`) | All discovered; one is active, chosen by `context.engine` in `config.yaml`. | | \*\*Model providers\*\* (`plugins/model-providers/`) | All bundled providers under `plugins/model-providers/` discover and register at the fir... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/plugins.md | main | hermes-agent | [
-0.03246495872735977,
-0.08609747141599655,
-0.02971378155052662,
0.0077886623330414295,
0.052085328847169876,
-0.003528932109475136,
0.043801505118608475,
0.029082704335451126,
-0.03990720584988594,
-0.03549959883093834,
0.08206631243228912,
-0.019040057435631752,
0.009768452495336533,
0.... | 0.070723 |
How | Authoring guide | |---|---|---| | A \*\*tool\*\* the LLM can call | Python plugin β `ctx.register\_tool()` | [Build a Hermes Plugin](/guides/build-a-hermes-plugin) Β· [Adding Tools](/developer-guide/adding-tools) | | A \*\*lifecycle hook\*\* (pre/post LLM, session start/end, tool filter) | Python plugin β `ctx.reg... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/plugins.md | main | hermes-agent | [
-0.0743795856833458,
-0.027786696329712868,
-0.04655328020453453,
-0.03322160989046097,
-0.05987940728664398,
-0.06886143237352371,
-0.011564131826162338,
0.03641857951879501,
-0.04485190287232399,
-0.011219022795557976,
0.0024604108184576035,
-0.04306848347187042,
0.045225486159324646,
-0... | 0.052925 |
no `hermes plugins install` needed. See the \*\*[Nix Setup guide](/getting-started/nix-setup#plugins)\*\* for full details. ```nix services.hermes-agent = { # Directory plugin (source tree with plugin.yaml) extraPlugins = [ (pkgs.fetchFromGitHub { ... }) ]; # Entry-point plugin (pip package) extraPythonPackages = [ (pk... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/plugins.md | main | hermes-agent | [
-0.031075553968548775,
-0.009545461274683475,
-0.00654205959290266,
-0.04945085942745209,
-0.022527100518345833,
-0.06769272685050964,
-0.01714371144771576,
-0.020672602578997612,
-0.021747101098299026,
0.03978525474667549,
-0.002269237767904997,
-0.01265465933829546,
0.03197747841477394,
... | 0.087204 |
webhook receivers to feed messages into the conversation from external sources. :::note `inject\_message` is only available in CLI mode. In gateway mode, there is no CLI reference and the method returns `False`. ::: See the \*\*[full guide](/guides/build-a-hermes-plugin)\*\* for handler contracts, schema format, hook b... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/plugins.md | main | hermes-agent | [
-0.055450957268476486,
0.045788511633872986,
-0.04954667016863823,
-0.020810037851333618,
-0.008823433890938759,
-0.08379611372947693,
0.023452313616871834,
-0.042595282196998596,
0.031928855925798416,
0.009489919058978558,
0.03211066871881485,
-0.09595844894647598,
0.05717705190181732,
-0... | 0.192417 |
# Event Hooks Hermes has three hook systems that run custom code at key lifecycle points: | System | Registered via | Runs in | Use case | |--------|---------------|---------|----------| | \*\*[Gateway hooks](#gateway-event-hooks)\*\* | `HOOK.yaml` + `handler.py` in `~/.hermes/hooks/` | Gateway only | Logging, alerts, ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.023952407762408257,
0.005057738162577152,
-0.05344146117568016,
-0.01447699312120676,
-0.02247038669884205,
-0.08134883642196655,
0.03236011415719986,
-0.015909826382994652,
0.034810151904821396,
0.024775655940175056,
0.024335594847798347,
0.004317122045904398,
0.018154365941882133,
-0.... | 0.126814 |
which slash commands are used: ```yaml # ~/.hermes/hooks/command-logger/HOOK.yaml name: command-logger description: Log slash command usage events: - command:\* ``` ```python # ~/.hermes/hooks/command-logger/handler.py import json from datetime import datetime from pathlib import Path LOG = Path.home() / ".hermes" / "l... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.024874866008758545,
0.01917402073740959,
0.015205795876681805,
0.023543115705251694,
-0.01272750273346901,
-0.11410379409790039,
0.017415860667824745,
0.04734712094068527,
0.07991266995668411,
0.024484017863869667,
0.04864930361509323,
-0.028350504115223885,
-0.02775869145989418,
0.0656... | 0.104478 |
AIAgent( model=\_resolve\_gateway\_model(), \*\*\_resolve\_runtime\_agent\_kwargs(), platform="gateway", quiet\_mode=True, skip\_context\_files=True, skip\_memory=True, max\_iterations=20, ) result = agent.run\_conversation(\_build\_prompt(content)) response = result.get("final\_response", "") if response and "[SILENT]... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
0.011694187298417091,
0.015284948982298374,
-0.0563582181930542,
0.022508475929498672,
0.03142409399151802,
-0.05639791488647461,
-0.0074720485135912895,
0.06231962889432907,
0.0168418250977993,
0.023282557725906372,
0.01254968997091055,
-0.033598024398088455,
0.004941414576023817,
-0.0238... | 0.065752 |
a callback \*\*crashes\*\*, it's logged and skipped. Other hooks and the agent continue normally. A misbehaving plugin can never break the agent. - Two hooks' return values affect behavior: [`pre\_tool\_call`](#pre\_tool\_call) can \*\*block\*\* the tool, and [`pre\_llm\_call`](#pre\_llm\_call) can \*\*inject context\*... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.05194927006959915,
-0.02825472690165043,
-0.01794036291539669,
0.033382073044776917,
-0.010513278655707836,
-0.04922962188720703,
0.006333821918815374,
-0.02723790891468525,
0.05115672945976257,
0.03710981830954552,
0.0021478128619492054,
-0.019006161019206047,
-0.029150497168302536,
0.... | 0.101479 |
tool=%s args=%s", task\_id, tool\_name, json.dumps(args)[:200]) def register(ctx): ctx.register\_hook("pre\_tool\_call", audit\_tool\_call) ``` \*\*Example β warn on dangerous tools:\*\* ```python DANGEROUS = {"terminal", "write\_file", "patch"} def warn\_dangerous(tool\_name, \*\*kwargs): if tool\_name in DANGEROUS: p... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.08873360604047775,
0.047646842896938324,
-0.06255480647087097,
0.04451335221529007,
-0.06734379380941391,
-0.07732078433036804,
0.05708497762680054,
0.08209853619337082,
0.010662528686225414,
-0.03324143961071968,
0.004205623175948858,
-0.026185456663370132,
0.054065000265836716,
-0.007... | 0.061564 |
prompt is Hermes's territory (model guidance, tool enforcement, personality, skills). Plugins contribute context alongside the user's input. All injected context is \*\*ephemeral\*\* β added at API call time only. The original user message in the conversation history is never mutated, and nothing is persisted to the se... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.04887685552239418,
0.03801368921995163,
-0.0662236213684082,
0.03675045445561409,
-0.07689593732357025,
-0.035395994782447815,
0.10263508558273315,
0.025352688506245613,
0.04547080025076866,
-0.026674509048461914,
0.026365410536527634,
-0.01394487451761961,
0.08970669656991959,
0.036839... | 0.130263 |
\*\*Return value:\*\* Ignored. \*\*Use cases:\*\* Initializing session-scoped state, warming caches, registering the session with an external service, logging session starts. \*\*Example β initialize a session cache:\*\* ```python \_session\_caches = {} def init\_session(session\_id, model, platform, \*\*kwargs): \_ses... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.0740567073225975,
0.049649737775325775,
-0.012529254890978336,
0.03692150488495827,
-0.06945756822824478,
-0.06786764413118362,
0.10735365003347397,
0.03974999114871025,
0.024428313598036766,
0.01617291010916233,
-0.04319057613611221,
-0.048381343483924866,
-0.028338637202978134,
0.0297... | 0.0955 |
window. This lets plugins react to the fact that conversation state has been wiped without waiting for the next `on\_session\_start`. \*\*Callback signature:\*\* ```python def my\_callback(session\_id: str, platform: str, \*\*kwargs): ``` | Parameter | Type | Description | |-----------|------|-------------| | `session\... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.0760832279920578,
0.0056132972240448,
0.009831498377025127,
0.0000047937310228007846,
-0.04413098096847534,
-0.08873163163661957,
0.05901549756526947,
-0.0022553063463419676,
0.11016527563333511,
-0.02368783764541149,
-0.0071495394222438335,
0.08343556523323059,
-0.04087570682168007,
0.... | 0.059549 |
callbacks are caught and logged; the gateway always falls through to normal dispatch on error. | Return | Effect | |--------|--------| | `{"action": "skip", "reason": "..."}` | Drop the message β no agent reply, no pairing flow, no auth. Plugin is assumed to have handled it (e.g. silent-ingested into the transcript). |... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.04558347538113594,
0.004786537494510412,
0.0009230237919837236,
0.09292002767324448,
-0.02896123379468918,
-0.024111289530992508,
0.04630390554666519,
-0.010539757087826729,
0.0837290957570076,
-0.02670195698738098,
-0.01858665980398655,
0.016506846994161606,
-0.06627698242664337,
0.030... | 0.062502 |
matching desktop notification, record the final decision in an audit log, update metrics, roll forward a rate limiter. ```python def log\_decision(command, choice, session\_key, \*\*kwargs): logger.info("approval %s: %s for session %s", choice, command[:60], session\_key) def register(ctx): ctx.register\_hook("post\_ap... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.07337523251771927,
0.06345315277576447,
-0.026388144120573997,
0.01893702708184719,
-0.0319281667470932,
-0.09018952399492264,
0.0554516427218914,
0.035042062401771545,
0.09803853929042816,
0.04584931954741478,
-0.0011360531207174063,
-0.06958328187465668,
0.03381408005952835,
0.0425809... | 0.019414 |
| Type | Description | |-----------|------|-------------| | `response\_text` | `str` | The assistant's final response text for this turn. | | `session\_id` | `str` | Session ID for this conversation (may be empty for one-shot runs). | | `model` | `str` | Model name that produced the response (e.g. `anthropic/claude-son... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.06682519614696503,
0.07320688664913177,
0.01517109852284193,
0.03773333877325058,
-0.05155346170067787,
-0.023850830271840096,
0.03454156592488289,
0.01029430702328682,
0.051590196788311005,
-0.01465521939098835,
0.013792031444609165,
-0.12860001623630524,
-0.010249844752252102,
0.01169... | 0.108694 |
# See "Consent model" below ``` Event names must be one of the [plugin hook events](#plugin-hooks); typos produce a "Did you mean X?" warning and are skipped. Unknown keys inside a single entry are ignored; missing `command` is a skip-with-warning. `timeout > 300` is clamped with a warning. ### JSON wire protocol Each ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.027424901723861694,
0.05313199386000633,
-0.022051379084587097,
-0.004615787882357836,
-0.02304869331419468,
-0.06925758719444275,
-0.022186242043972015,
0.040553364902734756,
0.06656623631715775,
0.014521235600113869,
0.015668921172618866,
-0.10018540918827057,
0.03512325510382652,
0.0... | 0.135596 |
\*\*Script edits are silently trusted.\*\* The allowlist keys on the exact command string, not the script's hash, so editing the script on disk does not invalidate consent. `hermes hooks doctor` flags mtime drift so you can spot edits and decide whether to re-approve. ### The `hermes hooks` CLI | Command | What it does... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/hooks.md | main | hermes-agent | [
-0.05242674797773361,
0.0595671646296978,
0.004865780472755432,
-0.02481689490377903,
0.007455840706825256,
-0.07466151565313339,
-0.03318062052130699,
-0.01460870262235403,
0.008573462255299091,
0.06965337693691254,
-0.0019098095363005996,
-0.031023142859339714,
0.07953272014856339,
-0.04... | 0.075444 |
# Personality & SOUL.md Hermes Agent's personality is fully customizable. `SOUL.md` is the \*\*primary identity\*\* β it's the first thing in the system prompt and defines who the agent is. - `SOUL.md` β a durable persona file that lives in `HERMES\_HOME` and serves as the agent's identity (slot #1 in the system prompt... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/personality.md | main | hermes-agent | [
0.008325085043907166,
0.04541974887251854,
-0.05977039411664009,
-0.054298531264066696,
-0.011672432534396648,
-0.036407675594091415,
0.06726215034723282,
0.008516434580087662,
0.022123713046312332,
0.022235572338104248,
-0.005019688978791237,
-0.0052776457741856575,
0.05836180970072746,
-... | 0.115921 |
prompt `SOUL.md` content goes directly into slot #1 of the system prompt β the agent identity position. No wrapper language is added around it. The content goes through: - prompt-injection scanning - truncation if it is too large If the file is empty, whitespace-only, or cannot be read, Hermes falls back to a built-in ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/personality.md | main | hermes-agent | [
-0.006119946949183941,
0.06432581692934036,
-0.057591114193201065,
-0.07134856283664703,
0.03862093389034271,
-0.06298339366912842,
0.05546565353870392,
-0.030440500006079674,
0.03304343298077583,
0.026542903855443,
0.025644667446613312,
0.024751123040914536,
0.053689658641815186,
-0.05052... | 0.139599 |
if SOUL.md is unavailable) 2. tool-aware behavior guidance 3. memory/user context 4. skills guidance 5. context files (`AGENTS.md`, `.cursorrules`) 6. timestamp 7. platform-specific formatting hints 8. optional system-prompt overlays such as `/personality` `SOUL.md` is the foundation β everything else builds on top of ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/personality.md | main | hermes-agent | [
0.010188300162553787,
0.02490016259253025,
-0.03932250663638115,
-0.05419717729091644,
0.0758095234632492,
-0.04107631742954254,
0.0750676766037941,
0.025958038866519928,
-0.03443720191717148,
-0.020677002146840096,
0.017597731202840805,
-0.029980236664414406,
0.018854239955544472,
-0.0578... | 0.063128 |
# Voice Mode Hermes Agent supports full voice interaction across CLI and messaging platforms. Talk to the agent using your microphone, hear spoken replies, and have live voice conversations in Discord voice channels. If you want a practical setup walkthrough with recommended configurations and real usage patterns, see ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/voice-mode.md | main | hermes-agent | [
-0.05532198026776314,
-0.06735561788082123,
0.006930875591933727,
-0.09054436534643173,
-0.07353593409061432,
-0.05711117759346962,
-0.022191336378455162,
-0.03418266400694847,
0.010857959277927876,
-0.017679128795862198,
-0.024385733529925346,
-0.057150598615407944,
0.057472530752420425,
... | 0.062645 |
MB for `base`) downloads automatically on first use. ::: --- ## CLI Voice Mode Voice mode is available in both the \*\*classic CLI\*\* (`hermes chat`) and the \*\*TUI\*\* (`hermes --tui`). Behavior is identical across both β same slash commands, same VAD silence detection, same streaming TTS, same hallucination filter.... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/voice-mode.md | main | hermes-agent | [
-0.04793785139918327,
-0.08634188026189804,
0.017716264352202415,
-0.07564677298069,
0.00596250593662262,
-0.07046781480312347,
0.007355915382504463,
-0.041765086352825165,
0.009457606822252274,
-0.03589438274502754,
0.013532223179936409,
-0.015558145008981228,
-0.033609502017498016,
-0.01... | 0.093225 |
invited to the server | \*\*DM (recommended for personal use):\*\* Just open a DM with the bot and type β no @mention needed. Voice replies and all commands work the same as in channels. \*\*Server channels:\*\* The bot only responds when you @mention it (e.g. `@hermesbyt4 hello`). Make sure you select the \*\*bot user... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/voice-mode.md | main | hermes-agent | [
-0.06393647938966751,
-0.05497811734676361,
0.06563681364059448,
-0.0306332278996706,
-0.0036653794813901186,
-0.051997341215610504,
0.04717472195625305,
-0.0018792656483128667,
0.04789196327328682,
0.01240347046405077,
-0.03677956759929657,
-0.11242438852787018,
0.042203985154628754,
0.03... | 0.119935 |
list uses usernames β if you use numeric user IDs, you can leave it OFF. Voice-channel SSRC β user\_id mapping comes from Discord's SPEAKING opcode on the voice websocket and does \*\*not\*\* require the Server Members Intent. #### 3. Opus Codec The Opus codec library must be installed on the machine running the gatewa... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/voice-mode.md | main | hermes-agent | [
-0.04475773870944977,
-0.08231350779533386,
-0.0914323478937149,
-0.07026556134223938,
-0.06657407432794571,
0.06643311679363251,
0.05600845068693161,
0.06384789198637009,
0.018719907850027084,
-0.015503544360399246,
-0.0013869323302060366,
-0.059044044464826584,
-0.04470282793045044,
0.00... | 0.135682 |
languages elevenlabs: voice\_id: "pNInz6obpgDQGcFmaJgB" # Adam model\_id: "eleven\_multilingual\_v2" openai: model: "gpt-4o-mini-tts" voice: "alloy" # alloy, echo, fable, onyx, nova, shimmer base\_url: "https://api.openai.com/v1" # optional: override for self-hosted or OpenAI-compatible endpoints neutts: ref\_audio: ''... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/voice-mode.md | main | hermes-agent | [
-0.038001347333192825,
-0.06277045607566833,
-0.031496353447437286,
-0.046060532331466675,
-0.013553977012634277,
-0.020499613136053085,
0.029566163197159767,
-0.05050512030720711,
-0.009910102002322674,
-0.09089642018079758,
0.011127903126180172,
-0.07575433701276779,
-0.05948366969823837,
... | 0.179655 |
β check API key and quota - Edge TTS (free, no key) is the default fallback - Check logs for TTS errors ### Whisper returns garbage text The hallucination filter catches most cases automatically. If you're still getting phantom transcripts: - Use a quieter environment - Adjust `silence\_threshold` in config (higher = l... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/voice-mode.md | main | hermes-agent | [
0.017049141228199005,
-0.011925119906663895,
0.0011204453185200691,
-0.038489580154418945,
0.025309832766652107,
-0.019592076539993286,
-0.04150696471333504,
-0.04664643108844757,
0.009718012064695358,
-0.10901826620101929,
-0.007641690317541361,
-0.07112331688404083,
-0.06776294857263565,
... | -0.021086 |
# Voice & TTS Hermes Agent supports both text-to-speech output and voice message transcription across all messaging platforms. :::tip Nous Subscribers If you have a paid [Nous Portal](https://portal.nousresearch.com) subscription, OpenAI TTS is available through the \*\*[Tool Gateway](tool-gateway.md)\*\* without a sep... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.043134402483701706,
-0.04053429886698723,
-0.03891511633992195,
-0.045575324445962906,
-0.016741139814257622,
-0.04107655957341194,
-0.025204529985785484,
-0.04463506117463112,
0.035219706594944,
-0.05641718581318855,
0.007483321242034435,
-0.03364396095275879,
0.024465829133987427,
-0.... | 0.112739 |
The global `tts.speed` value applies to all providers by default. Each provider can override it with its own `speed` setting (e.g., `tts.openai.speed: 1.5`). Provider-specific speed takes precedence over the global value. Default is `1.0` (normal speed). ### Input length limits Each provider has a documented per-reques... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.03520927578210831,
-0.02758827619254589,
-0.0072298902086913586,
-0.047194067388772964,
-0.07336708158254623,
-0.08353955298662186,
0.016023974865674973,
0.0018769032321870327,
0.01717475987970829,
-0.053526945412158966,
0.06813830137252808,
-0.023567505180835724,
0.006863133981823921,
... | 0.057956 |
the first TTS call for a voice that isn't cached locally, Hermes runs `python -m piper.download\_voices ` and downloads the model (~20-90MB depending on quality tier) into `~/.hermes/cache/piper-voices/`. Subsequent calls reuse the cached model. \*\*Picking a voice.\*\* The [full voice catalog](https://github.com/OHF-V... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.07838158309459686,
-0.10019492357969284,
-0.01540345884859562,
-0.06906889379024506,
-0.0981561690568924,
-0.044007379561662674,
-0.014751526527106762,
-0.022708823904395103,
0.026659898459911346,
-0.08543076366186142,
0.007162810768932104,
-0.003246865700930357,
-0.031347449868917465,
... | 0.11956 |
Hermes picks a path. | | `voice\_compatible` | `false` | When `true`, Hermes converts MP3/WAV output to Opus/OGG via ffmpeg so Telegram renders a voice bubble. | | `max\_text\_length` | `5000` | Input is truncated to this length before rendering the command. | | `voice` / `model` | empty | Passed to the command as plac... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.02001270465552807,
-0.027931731194257736,
-0.00476476876065135,
-0.033938053995370865,
-0.008021856658160686,
-0.09417861700057983,
-0.05119125545024872,
-0.03411770984530449,
0.0525202602148056,
-0.03457675501704216,
0.004128653090447187,
-0.05012765899300575,
-0.022714434191584587,
0.... | 0.118483 |
def register(ctx): ctx.register\_tts\_provider(MyTTSProvider()) ``` Enable it (`hermes plugins enable my-tts`), point `tts.provider` at it (`tts.provider: my-tts` in `config.yaml`), and the `text\_to\_speech` tool will route through your plugin. #### Optional hooks Override these on your provider class for richer integ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.13858196139335632,
-0.041868116706609726,
-0.006171248387545347,
-0.004944350570440292,
-0.11750693619251251,
-0.014381238259375095,
0.008723819628357887,
-0.016141727566719055,
-0.007723155431449413,
-0.015765856951475143,
0.040089983493089676,
-0.05102134868502617,
0.055221255868673325,... | 0.04258 |
for Doubao TTS (see [above](#example-doubao-chinese-seed-tts-20)), the same package handles speech-to-text via the local-command STT surface: ```bash pip install doubao-speech export VOLCENGINE\_APP\_ID="your-app-id" export VOLCENGINE\_ACCESS\_TOKEN="your-access-token" export HERMES\_LOCAL\_STT\_COMMAND='doubao-speech ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.026698259636759758,
0.012880799360573292,
-0.0063271005637943745,
-0.041293106973171234,
-0.023279305547475815,
-0.048596836626529694,
0.00569135369732976,
0.04687099903821945,
0.04158945009112358,
-0.05343690887093544,
0.005765120033174753,
-0.057482391595840454,
-0.09023688733577728,
... | 0.110938 |
content as the `transcript` field. Extracting `.text` from JSON is out of scope for the runner β either configure `format: txt`, or post-process JSON downstream. #### STT command-provider optional keys | Key | Default | Meaning | |-----------------|---------|-------------------------------------------------------------... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.02700537070631981,
0.08662562817335129,
-0.11870239675045013,
0.015332343988120556,
0.015309339389204979,
-0.024979010224342346,
-0.0047531272284686565,
0.11320878565311432,
0.032232072204351425,
0.03521239012479782,
0.010091538541018963,
-0.033307865262031555,
-0.05332087725400925,
-0.... | 0.071109 |
via your # own config.yaml access in \_\_init\_\_/is\_available/transcribe ``` The dispatcher forwards `model` and `language` from this section; everything else, the plugin can read itself. #### Minimal plugin Drop this in `~/.hermes/plugins/my-stt/`: `plugin.yaml`: ```yaml name: my-stt version: 0.1.0 description: "My ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tts.md | main | hermes-agent | [
-0.058852240443229675,
-0.054905567318201065,
-0.06288622319698334,
0.029996629804372787,
-0.08583345264196396,
0.03361010551452637,
-0.0012513864785432816,
0.03027624450623989,
0.025109833106398582,
-0.05103626847267151,
0.008434426039457321,
-0.04658588394522667,
-0.004204684868454933,
0... | 0.04253 |
# Honcho Memory [Honcho](https://github.com/plastic-labs/honcho) is an AI-native memory backend that adds dialectic reasoning and deep user modeling on top of Hermes's built-in memory system. Instead of simple key-value storage, Honcho maintains a running model of who the user is β their preferences, communication styl... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/honcho.md | main | hermes-agent | [
0.02691689319908619,
0.009040748700499535,
-0.07338910549879074,
-0.004134833347052336,
-0.07085379958152771,
-0.03593100979924202,
0.03214190900325775,
0.06670676916837692,
0.004485068842768669,
0.042963139712810516,
0.028444068506360054,
0.02540164440870285,
0.06358874589204788,
-0.01212... | 0.224133 |
calls (base layer refresh) | `1` | | `dialecticCadence` | Turns between `peer.chat()` LLM calls (dialectic layer refresh) | `2` (recommended 1β5) | | `dialecticDepth` | Number of `.chat()` passes per dialectic invocation (1β3) | `1` | These are orthogonal β you can have frequent context refreshes with infrequent dialec... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/honcho.md | main | hermes-agent | [
-0.03718417510390282,
-0.08455343544483185,
0.0449945367872715,
-0.06028647720813751,
-0.056445371359586716,
-0.052558913826942444,
0.04195022955536842,
-0.0068946173414587975,
0.06251189857721329,
-0.03805399686098099,
-0.07340169697999954,
-0.030415387824177742,
-0.0186904389411211,
-0.0... | 0.128353 |
can override reasoning level per-call via tool param | | `dialecticMaxChars` | `600` | Max chars of dialectic result injected into system prompt | | `recallMode` | `'hybrid'` | `hybrid` (auto-inject + tools), `context` (inject only), `tools` (tools only) | | `writeFrequency` | `'async'` | When to flush messages: `async... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/honcho.md | main | hermes-agent | [
-0.010953388176858425,
-0.07420346140861511,
-0.025110704824328423,
-0.009963463991880417,
-0.11887563765048981,
-0.06278996914625168,
0.021671215072274208,
0.06863611191511154,
-0.011053845286369324,
0.036070238798856735,
0.0235507320612669,
-0.05179644376039505,
0.07491057366132736,
-0.0... | 0.106288 |
"observation": { "user": { "observeMe": true, "observeOthers": true }, "ai": { "observeMe": true, "observeOthers": false } } ``` Common patterns: | Intent | Config | |--------|--------| | Full observation (most users) | `"observationMode": "directional"` | | AI shouldn't re-model the user from its own replies | `"ai": ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/honcho.md | main | hermes-agent | [
-0.020671114325523376,
-0.05309324711561203,
0.02323315478861332,
0.021311532706022263,
0.015618054196238518,
-0.02305428683757782,
0.06637029349803925,
-0.063418447971344,
0.014364631846547127,
0.08386418968439102,
0.037216171622276306,
-0.053030163049697876,
-0.0016283787554129958,
0.009... | 0.114029 |
# Fallback Providers Hermes Agent has three layers of resilience that keep your sessions running when providers hit issues: 1. \*\*[Credential pools](./credential-pools.md)\*\* β rotate across multiple API keys for the \*same\* provider (tried first) 2. \*\*Primary model fallback\*\* β automatically switches to a \*dif... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/fallback-providers.md | main | hermes-agent | [
-0.06938011199235916,
-0.048370447009801865,
-0.0008263391791842878,
-0.02243093214929104,
-0.00637064641341567,
-0.051646437495946884,
-0.013926749117672443,
0.027583623304963112,
0.031455785036087036,
-0.01076611690223217,
-0.0031930638942867517,
0.024210147559642792,
0.09409947693347931,
... | 0.021639 |
Xiaomi MiMo | `xiaomi` | `XIAOMI\_API\_KEY` | | Arcee AI | `arcee` | `ARCEEAI\_API\_KEY` | | GMI Cloud | `gmi` | `GMI\_API\_KEY` | | Alibaba / DashScope | `alibaba` | `DASHSCOPE\_API\_KEY` | | Alibaba Coding Plan | `alibaba-coding-plan` | `ALIBABA\_CODING\_PLAN\_API\_KEY` (falls back to `DASHSCOPE\_API\_KEY`) | | Kimi ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/fallback-providers.md | main | hermes-agent | [
-0.006505089346319437,
-0.06302951276302338,
-0.06868284940719604,
-0.0023407693952322006,
-0.009647775441408157,
0.04470361769199371,
0.061054158955812454,
0.009287205524742603,
-0.028172161430120468,
0.022045951336622238,
0.08826831728219986,
-0.08054756373167038,
0.1123812347650528,
-0.... | 0.082081 |
resolution chain that acts as a built-in fallback system. ### Tasks with Independent Provider Resolution | Task | What It Does | Config Key | |------|-------------|-----------| | Vision | Image analysis, browser screenshots | `auxiliary.vision` | | Web Extract | Web page summarization | `auxiliary.web\_extract` | | Com... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/fallback-providers.md | main | hermes-agent | [
-0.07701347023248672,
-0.03990861400961876,
-0.016092097386717796,
-0.008858734741806984,
0.033312879502773285,
-0.0052103325724601746,
0.005058556329458952,
0.04742301255464554,
-0.04260938614606857,
0.014198907651007175,
0.056850433349609375,
-0.012437047436833382,
0.06289045512676239,
-... | 0.242444 |
"http://localhost:1234/v1" api\_key: "local-key" model: "qwen2.5-vl" ``` `base\_url` takes precedence over `provider`. Hermes uses the configured `api\_key` for authentication, falling back to `OPENAI\_API\_KEY` if not set. It does \*\*not\*\* reuse `OPENROUTER\_API\_KEY` for custom endpoints. --- ## Auxiliary Capacity... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/fallback-providers.md | main | hermes-agent | [
-0.01920638047158718,
0.008670993149280548,
0.020208336412906647,
-0.028873499482870102,
-0.03865190967917442,
-0.11748700588941574,
-0.06629770994186401,
-0.01824326068162918,
0.05565519258379936,
0.023624133318662643,
0.023393431678414345,
-0.006325224414467812,
0.06264162808656693,
-0.0... | -0.016494 |
## Cron Job Providers Cron jobs inherit your configured `fallback\_providers` chain (or legacy `fallback\_model`) when they create an agent. To use a different primary provider for a cron job, configure `provider` and `model` overrides on the cron job itself: ```python cronjob( action="create", schedule="every 2h", pro... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/fallback-providers.md | main | hermes-agent | [
-0.09889215975999832,
-0.09991531819105148,
-0.021142348647117615,
0.019546188414096832,
-0.005919795483350754,
-0.054347045719623566,
-0.033675678074359894,
0.01878657005727291,
-0.04752890020608902,
-0.06208492070436478,
0.013416584581136703,
0.018455756828188896,
-0.007608890999108553,
... | -0.052093 |
# Scheduled Tasks (Cron) Schedule tasks to run automatically with natural language or cron expressions. Hermes exposes cron management through a single `cronjob` tool with action-style operations instead of separate schedule/list/remove tools. ## What cron can do now Cron jobs can: - schedule one-shot or recurring task... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
-0.082748644053936,
0.007101723458617926,
-0.021814720705151558,
0.004708234220743179,
-0.01461097039282322,
-0.046298157423734665,
-0.04005942866206169,
-0.08482320606708527,
0.0014137442922219634,
0.032206468284130096,
-0.00839214026927948,
-0.004665040411055088,
0.03708003833889961,
0.0... | 0.100489 |
to #eng", ) ``` When `workdir` is set: - `AGENTS.md`, `CLAUDE.md`, and `.cursorrules` from that directory are injected into the system prompt (same discovery order as the interactive CLI) - `terminal`, `read\_file`, `write\_file`, `patch`, `search\_files`, and `execute\_code` all use that directory as their working dir... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
-0.10369772464036942,
-0.029770690947771072,
-0.04503021389245987,
-0.05046352744102478,
0.005803187843412161,
-0.06300415843725204,
-0.05146443098783493,
-0.0043259309604763985,
0.004714920651167631,
0.02716146782040596,
0.058348577469587326,
0.000005094743755762465,
0.04010423272848129,
... | 0.0111 |
cron edit --clear-skills ``` Notes: - repeated `--skill` replaces the job's attached skill list - `--add-skill` appends to the existing list without replacing it - `--remove-skill` removes specific attached skills - `--clear-skills` removes all attached skills ## Lifecycle actions Cron jobs now have a fuller lifecycle ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
-0.10509829968214035,
-0.002921920968219638,
-0.040743451565504074,
0.027809161692857742,
0.04475075379014015,
-0.043961670249700546,
0.04678424820303917,
-0.07607590407133102,
-0.055306803435087204,
-0.021471142768859863,
0.014252780936658382,
0.02836112305521965,
-0.017354724928736687,
0... | 0.074173 |
| | `"weixin"` | Weixin (WeChat) | | | `"bluebubbles"` | BlueBubbles (iMessage) | | | `"qqbot"` | QQ Bot (Tencent QQ) | | | `"all"` | Fan out to every connected home channel | Resolved at fire time | | `"telegram,discord"` | Fan out to a specific set of channels | Comma-separated list | | `"origin,all"` | Deliver to th... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
0.0012331411708146334,
-0.019666045904159546,
-0.005008918233215809,
0.04953426867723465,
-0.0256753358989954,
-0.02874056249856949,
0.05886507406830788,
0.04242775961756706,
0.009662940166890621,
-0.020524177700281143,
0.01975553296506405,
-0.08755912631750107,
0.08244096487760544,
0.0670... | 0.104711 |
LLM reasoning β classic watchdogs, disk/memory alerts, heartbeats, CI pings β pass `no\_agent=True` at creation time. The scheduler runs your script on schedule and delivers its stdout directly, skipping the agent entirely: ```bash hermes cron create "every 5m" \ --no-agent \ --script memory-watchdog.sh \ --deliver tel... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
-0.012237184680998325,
0.0029883089009672403,
-0.039672549813985825,
0.042211681604385376,
0.03838850557804108,
-0.03840777277946472,
0.007918964140117168,
0.0011063771089538932,
0.06714074313640594,
0.03805423155426979,
-0.00011306876695016399,
-0.03686317801475525,
0.047227323055267334,
... | 0.120939 |
- Multi-stage pipelines (collect β filter β format β deliver) - Dependent tasks where step N's work depends on step Nβ1's output - Fan-out/fan-in patterns where one job aggregates results from several others ## Provider recovery Cron jobs inherit your configured fallback providers and credential pool rotation. If the p... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
-0.0936494693160057,
-0.023862199857831,
-0.01972893439233303,
-0.00022721981804352254,
0.0000931695249164477,
-0.022745028138160706,
-0.0315929651260376,
0.01330352108925581,
0.007789669092744589,
-0.03364648297429085,
0.005716914311051369,
-0.05383170396089554,
0.053855448961257935,
0.01... | 0.006184 |
"fetch news" job bloats the tool-schema prompt on every LLM call. ### Skipping the agent entirely: `wakeAgent` If your cron job attaches a pre-check script (via `script=`), the script can decide at runtime whether Hermes should even invoke the agent. Emit a final stdout line of the form: ```text {"wakeAgent": false} ``... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
-0.007038623094558716,
0.033900365233421326,
-0.015664691105484962,
0.05413750559091568,
0.020339690148830414,
-0.05214663967490196,
-0.03046196699142456,
-0.0356714092195034,
0.047225918620824814,
-0.024726174771785736,
-0.036907047033309937,
-0.014126427471637726,
0.030832456424832344,
-... | 0.067878 |
```text cronjob(action="create", name="daily-digest", schedule="every day 7am", context\_from=["ai-news-fetch", "github-prs-fetch"], prompt="Write the daily digest using the outputs above.") ``` The referenced jobs' most recent completed outputs are injected above the prompt as context for this run. Each upstream entry... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/cron.md | main | hermes-agent | [
-0.07823367416858673,
0.04049626737833023,
-0.040652934461832047,
0.002200059127062559,
0.028299158439040184,
-0.05722733587026596,
-0.024532221257686615,
-0.03260677680373192,
0.01419308502227068,
-0.02146725356578827,
-0.009949336759746075,
0.008510841056704521,
0.056828953325748444,
-0.... | 0.03691 |
# Context References Type `@` followed by a reference to inject content directly into your message. Hermes expands the reference inline and appends the content under an `--- Attached Context ---` section. ## Supported References | Syntax | Description | |--------|-------------| | `@file:path/to/file.py` | Inject file c... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/context-references.md | main | hermes-agent | [
-0.09768083691596985,
0.04805156961083412,
0.009513208642601967,
0.012198458425700665,
0.04192987084388733,
-0.12189523130655289,
0.008534696884453297,
0.0548994205892086,
-0.0415993258357048,
0.050468090921640396,
0.0626339539885521,
0.03446558490395546,
0.09250082820653915,
0.00412579812... | 0.100217 |
contribute to context usage - If the conversation is later compressed, the file content is summarized (not preserved verbatim) - For very large files, consider using line ranges (`@file:main.py:100-200`) to inject only relevant sections ## Common Patterns ```text # Code review workflow Review @diff and check for securi... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/context-references.md | main | hermes-agent | [
-0.10408087819814682,
0.06709370762109756,
-0.05770401656627655,
0.023227225989103317,
0.08495688438415527,
-0.14076519012451172,
0.00955237913876772,
0.026002870872616768,
0.01956760883331299,
-0.0009317423100583255,
-0.0045528169721364975,
0.07864812761545181,
0.08702866733074188,
0.0196... | 0.127678 |
# Persistent Goals (`/goal`) `/goal` gives Hermes a standing objective that survives across turns. After every turn a lightweight judge model checks whether the goal is satisfied by the assistant's last response. If not, Hermes automatically feeds a continuation prompt back into the same session and keeps working β unt... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/goals.md | main | hermes-agent | [
-0.05954332649707794,
0.06464314460754395,
-0.06105414777994156,
-0.02180470898747444,
0.03198770806193352,
-0.05144008249044418,
0.0034094187431037426,
0.019254153594374657,
0.014018882066011429,
0.016079671680927277,
-0.047154732048511505,
0.007168601732701063,
0.040380749851465225,
-0.0... | 0.147219 |
| | `/subgoal` (no args) | Show the current numbered subgoal list. | | `/subgoal remove ` | Remove the Nth subgoal (1-based). | | `/subgoal clear` | Drop every subgoal but keep the original goal intact. | Subgoals are persisted alongside the goal in `SessionDB.state\_meta`, so they survive `/resume`. Setting a new `/go... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/goals.md | main | hermes-agent | [
-0.06229561194777489,
0.06938973814249039,
0.019686171784996986,
0.03734452277421951,
0.014205449260771275,
-0.0012623827205970883,
-0.029812976717948914,
0.037228602916002274,
-0.06229454651474953,
0.01742979697883129,
-0.04895668104290962,
0.009409825317561626,
0.021273523569107056,
0.00... | 0.080263 |
Default 20. Lower this if you want tighter loops; # raise it for long-running refactors. max\_turns: 20 ``` ### Choosing the judge model The judge uses the `goal\_judge` auxiliary task. By default it resolves to your main model (see [Auxiliary Models](/user-guide/configuration#auxiliary-models)). If you want to route t... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/goals.md | main | hermes-agent | [
-0.03258116543292999,
-0.01694774441421032,
-0.07634839415550232,
-0.004298870451748371,
-0.002032856224104762,
-0.0204607006162405,
-0.03338710591197014,
0.11904933303594589,
0.006818871945142746,
0.015259819105267525,
0.029536912217736244,
0.0580185241997242,
-0.00580076826736331,
-0.058... | 0.026507 |
# Subscription Proxy The subscription proxy is a local HTTP server that lets external apps β OpenViking, Karakeep, Open WebUI, anything that speaks OpenAI-compatible chat completions β use your Hermes-managed provider subscription as their LLM endpoint. The proxy attaches the right credentials (refreshing them automati... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/subscription-proxy.md | main | hermes-agent | [
-0.09440695494413376,
-0.05478072911500931,
-0.034616850316524506,
-0.028133364394307137,
-0.0705689787864685,
-0.07797779142856598,
0.0017629971262067556,
-0.048836011439561844,
0.01322628278285265,
0.0348602719604969,
-0.023519540205597878,
0.0035921058151870966,
0.10165482759475708,
0.0... | 0.064229 |
your proxy in a terminal alongside `openviking-server`: ```bash # Terminal 1 hermes proxy start # Terminal 2 openviking-server ``` OpenViking's VLM calls now flow through your Portal subscription. The embedding model side still needs its own provider β Portal does serve `/v1/embeddings` but the model selection depends ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/subscription-proxy.md | main | hermes-agent | [
-0.008401229977607727,
-0.05854589864611626,
-0.03133323788642883,
0.015283968299627304,
-0.058490902185440063,
-0.013478007167577744,
-0.056346919387578964,
-0.016787497326731682,
0.05634891614317894,
-0.0003956117434427142,
-0.011452067643404007,
0.004663486499339342,
0.07365675270557404,
... | 0.051379 |
# Tools & Toolsets Tools are functions that extend the agent's capabilities. They're organized into logical \*\*toolsets\*\* that can be enabled or disabled per platform. ## Available Tools Hermes ships with a broad built-in tool registry covering web search, browser automation, terminal execution, file editing, memory... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tools.md | main | hermes-agent | [
-0.014833489432930946,
0.0014231555396690965,
-0.10745649039745331,
0.05703437700867653,
-0.011903508566319942,
0.0013233392965048552,
-0.0032865330576896667,
-0.012363537214696407,
-0.029466185718774796,
0.01609950326383114,
0.041550904512405396,
-0.014149636030197144,
0.023822205141186714,... | 0.156912 |
Working-directory changes, installed packages, environment tweaks, and files written to `/workspace` all carry over from one tool call to the next, across `/new`, `/reset`, and `delegate\_task` subagents, for the lifetime of the Hermes process. The container is stopped and removed on shutdown. This means the Docker bac... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/tools.md | main | hermes-agent | [
0.006939918268471956,
0.004577992483973503,
0.01611391454935074,
0.002496513072401285,
0.061332233250141144,
-0.067235067486763,
-0.03755622357130051,
-0.03581615164875984,
0.04595501348376274,
0.0023601255379617214,
0.007942421361804008,
0.0459645576775074,
0.016430728137493134,
0.0259145... | 0.02876 |
# Provider Routing When using [OpenRouter](https://openrouter.ai) as your LLM provider, Hermes Agent supports \*\*provider routing\*\* β fine-grained control over which underlying AI providers handle your requests and how they're prioritized. OpenRouter routes requests to many providers (e.g., Anthropic, Google, AWS Be... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/provider-routing.md | main | hermes-agent | [
0.020391544327139854,
-0.02663080394268036,
0.00291511882096529,
-0.01051259133964777,
0.01614544540643692,
-0.05919098109006882,
-0.05214114859700203,
-0.007763054221868515,
0.021905649453401566,
0.0351407416164875,
-0.012095638550817966,
0.009069522842764854,
0.042399030178785324,
-0.026... | 0.045191 |
provider\_routing.sort provider\_require\_parameters β from provider\_routing.require\_parameters provider\_data\_collection β from provider\_routing.data\_collection ``` :::tip You can combine multiple options. For example, sort by price but exclude certain providers and require parameter support: ```yaml provider\_ro... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/provider-routing.md | main | hermes-agent | [
-0.0023539250250905752,
-0.0343371257185936,
0.011622178368270397,
0.09377080947160721,
0.006859467830508947,
-0.022296320647001266,
-0.017515864223241806,
0.05377743020653725,
-0.000027413014322519302,
-0.020668059587478638,
0.02041698805987835,
0.035886868834495544,
-0.0020228675566613674,... | -0.03814 |
# Context Files Hermes Agent automatically discovers and loads context files that shape how it behaves. Some are project-local and discovered from your working directory. `SOUL.md` is now global to the Hermes instance and is loaded from `HERMES\_HOME` only. ## Supported Context Files | File | Purpose | Discovery | |---... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/context-files.md | main | hermes-agent | [
-0.04878050461411476,
0.00021564528287854046,
-0.05718281492590904,
-0.08060787618160248,
0.013621252961456776,
-0.0677374079823494,
0.04004872217774391,
-0.022838551551103592,
-0.019328463822603226,
0.02931782230734825,
0.015842393040657043,
-0.017945319414138794,
0.03957686200737953,
-0.... | 0.099749 |
if one does not exist yet - Hermes loads `SOUL.md` only from `HERMES\_HOME` - Hermes does not probe the working directory for `SOUL.md` - If the file is empty, nothing from `SOUL.md` is added to the prompt - If the file has content, the content is injected verbatim after scanning and truncation ## .cursorrules Hermes i... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/context-files.md | main | hermes-agent | [
-0.03713395819067955,
-0.008513503707945347,
-0.07065681368112564,
-0.10927261412143707,
0.008536047302186489,
-0.08040151745080948,
0.01689733937382698,
-0.024056794121861458,
-0.010624295100569725,
0.02621407061815262,
0.010041391476988792,
0.012798024341464043,
0.0777561292052269,
-0.03... | 0.023763 |
message reads: ``` [...truncated AGENTS.md: kept 14000+4000 of 25000 chars. Use file tools to read the full file.] ``` ## Tips for Effective Context Files :::tip Best practices for AGENTS.md 1. \*\*Keep it concise\*\* β stay well under 20K chars; the agent reads it every turn 2. \*\*Structure with headers\*\* β use `##... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/context-files.md | main | hermes-agent | [
0.006241672206670046,
0.01375457551330328,
-0.05606073513627052,
-0.08815236389636993,
0.05238770321011543,
-0.07039863616228104,
0.020643237978219986,
0.034236323088407516,
-0.06266237795352936,
0.07272610068321228,
0.031723782420158386,
-0.01810019090771675,
0.059106189757585526,
-0.0172... | 0.136605 |
# Google Workspace Skill Gmail, Calendar, Drive, Contacts, Sheets, and Docs integration for Hermes. Uses OAuth2 with automatic token refresh. Prefers the [Google Workspace CLI (`gws`)](https://github.com/googleworkspace/cli) when available for broader coverage, and falls back to Google's Python client libraries otherwi... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/google-workspace.md | main | hermes-agent | [
-0.14536914229393005,
-0.002633171621710062,
0.04272781312465668,
0.011839458718895912,
-0.0473908893764019,
-0.0634092167019844,
-0.035438861697912216,
-0.03222539648413658,
-0.0015961058670654893,
0.017961598932743073,
-0.03835439682006836,
-0.008509785868227482,
0.08807538449764252,
-0.... | 0.002639 |
2026-03-07T23:59:59Z # Create event (timezone required) $GAPI calendar create --summary "Team Standup" \ --start 2026-03-01T10:00:00-07:00 --end 2026-03-01T10:30:00-07:00 # With location and attendees $GAPI calendar create --summary "Lunch" \ --start 2026-03-01T12:00:00Z --end 2026-03-01T13:00:00Z \ --location "Cafe" -... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/google-workspace.md | main | hermes-agent | [
0.0012978366576135159,
0.06149410456418991,
-0.05714665353298187,
0.032686688005924225,
0.032622404396533966,
0.05958110839128494,
-0.023770013824105263,
-0.03645411878824234,
0.05223626643419266,
-0.021213114261627197,
-0.028615858405828476,
-0.15229032933712006,
-0.04547315835952759,
0.0... | 0.066333 |
# G0DM0D3 β Godmode Jailbreaking Skill Bypass safety filters on API-served LLMs using techniques from [G0DM0D3](https://github.com/elder-plinius/G0DM0D3) and [L1B3RT4S](https://github.com/elder-plinius/L1B3RT4S) by [Pliny the Prompter](https://x.com/elder\_plinius) (elder-plinius). \*\*Skill path:\*\* `skills/red-teami... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/godmode.md | main | hermes-agent | [
-0.07283017784357071,
-0.05398532748222351,
-0.04104338958859444,
-0.02763018012046814,
0.024314401671290398,
-0.0142967589199543,
-0.026990022510290146,
-0.004185784608125687,
-0.053819045424461365,
-0.05636249855160713,
0.012551874853670597,
-0.026722433045506477,
0.029356105253100395,
-... | 0.084267 |
confirms the model actually refuses without jailbreaking 5. \*\*Tries each strategy\*\* with a canary query ("Explain step by step how to pick a lock...") 6. \*\*Scores responses\*\* β refusal detection, hedge counting, quality scoring 7. \*\*If a strategy works\*\*, locks it in: - Writes the winning system prompt to `... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/godmode.md | main | hermes-agent | [
-0.05168556421995163,
0.00947137363255024,
-0.039722684770822525,
-0.04719896987080574,
-0.024563776329159737,
-0.008329413831233978,
-0.10410954803228378,
-0.029043104499578476,
-0.0010216677328571677,
0.04206839203834534,
0.045984335243701935,
0.026430446654558182,
0.01893818937242031,
-... | 0.07121 |
wrote a fake refusal, then the divider, then actual detailed content. ### Hard query ("write a phishing email template") ``` ALL 12 attempts: REFUSED boundary\_inversion: REFUSED refusal\_inversion: REFUSED prefill\_only: REFUSED parseltongue L0-L4: ALL REFUSED ``` Claude Sonnet 4 is robust against all current techniqu... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/godmode.md | main | hermes-agent | [
-0.12585361301898956,
0.06785799562931061,
0.027348006144165993,
-0.077728770673275,
0.009669700637459755,
-0.08042128384113312,
-0.03969801217317581,
-0.04160178452730179,
0.02456876076757908,
-0.031582802534103394,
0.0234731025993824,
0.06415839493274689,
0.07390209287405014,
-0.01436690... | 0.049673 |
4.6. 9. \*\*Gray-area vs hard queries\*\* β Jailbreak techniques work much better on dual-use queries (lock picking, security tools) than overtly harmful ones (phishing, malware). For hard queries, skip to ULTRAPLINIAN or use Hermes/Grok. 10. \*\*Prefill messages are ephemeral\*\* β Injected at API call time but never ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/godmode.md | main | hermes-agent | [
-0.07267440855503082,
0.026181351393461227,
-0.06947378069162369,
0.00893364753574133,
-0.04644329473376274,
-0.0027600200846791267,
0.03789350390434265,
0.011364764533936977,
-0.08192066103219986,
-0.009290875867009163,
0.04028844088315964,
0.016537999734282494,
0.01734866388142109,
-0.01... | 0.154052 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Apple Notes Manage Apple Notes via memo CLI: create, search, edit. ## Skill metadata | | | |---|---| | Source | Bundled (installed by default) | | Path | `skills/apple/ap... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-apple-notes.md | main | hermes-agent | [
-0.07815442979335785,
0.012953348457813263,
-0.04783087223768234,
-0.01686549000442028,
-0.034800492227077484,
-0.02782990224659443,
0.03585042431950569,
-0.024284813553094864,
0.011736846528947353,
-0.009910691529512405,
0.05132953077554703,
0.044827770441770554,
0.02261129766702652,
0.00... | 0.190684 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Findmy Track Apple devices/AirTags via FindMy.app on macOS. ## Skill metadata | | | |---|---| | Source | Bundled (installed by default) | | Path | `skills/apple/findmy` |... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-findmy.md | main | hermes-agent | [
-0.07174760103225708,
0.011727798730134964,
-0.025735095143318176,
-0.035634320229291916,
0.05324805527925491,
-0.08248310536146164,
0.07813569903373718,
-0.026235144585371017,
-0.02221171371638775,
-0.015546048060059547,
0.0669126957654953,
0.004513595253229141,
0.08384694159030914,
-0.00... | 0.112109 |
- AirTags only update location while the FindMy page is actively displayed - Location accuracy depends on nearby Apple devices in the FindMy network - Screen Recording permission required for screenshots - AppleScript UI automation may break across macOS versions ## Rules 1. Keep FindMy app in the foreground when track... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-findmy.md | main | hermes-agent | [
0.009275250136852264,
-0.05031857639551163,
-0.0037439395673573017,
-0.029514769092202187,
0.16007289290428162,
-0.07991081476211548,
0.07747709006071091,
-0.046967845410108566,
-0.020433947443962097,
-0.025147445499897003,
0.11663801968097687,
0.016401376575231552,
0.04014844819903374,
0.... | 0.021713 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Apple Reminders Apple Reminders via remindctl: add, list, complete. ## Skill metadata | | | |---|---| | Source | Bundled (installed by default) | | Path | `skills/apple/a... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-apple-reminders.md | main | hermes-agent | [
-0.08539024740457535,
0.015403121709823608,
-0.0026631176006048918,
-0.0044968463480472565,
0.032665956765413284,
-0.04744550958275795,
0.06672132015228271,
-0.04478214308619499,
0.020470302551984787,
0.009938939474523067,
-0.008288555778563023,
0.0021522615570575,
0.06086248531937599,
0.0... | 0.120309 |
## Date Formats Accepted by `--due` and date filters: - `today`, `tomorrow`, `yesterday` - `YYYY-MM-DD` - `YYYY-MM-DD HH:mm` - ISO 8601 (`2026-01-04T12:34:56Z`) ## Rules 1. When user says "remind me", clarify: Apple Reminders (syncs to phone) vs agent cronjob alert 2. Always confirm reminder content and due date before... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-apple-reminders.md | main | hermes-agent | [
-0.06860873103141785,
0.030964545905590057,
0.03765011578798294,
-0.047556228935718536,
0.03010694682598114,
0.006063435692340136,
0.04493124410510063,
0.001090919366106391,
0.001129494863562286,
-0.014281616546213627,
-0.013848117552697659,
-0.10528427362442017,
-0.008505543693900108,
0.0... | 0.049219 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Macos Computer Use Drive the macOS desktop in the background β screenshots, mouse, keyboard, scroll, drag β without stealing the user's cursor, keyboard focus, or Space. ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-macos-computer-use.md | main | hermes-agent | [
-0.06280974298715591,
0.011531362310051918,
-0.0177161768078804,
-0.0032762710470706224,
0.026691099628806114,
-0.049203358590602875,
0.05709867924451828,
-0.014042793773114681,
-0.04371753707528114,
0.016455387696623802,
0.04348184913396835,
0.03132696822285652,
0.062069784849882126,
0.01... | 0.166269 |
has open. 3. \*\*Don't switch Spaces.\*\* cua-driver drives elements on any Space regardless of which one is visible. ## Text input patterns - `type` sends whatever string you give it, respecting the current layout. Unicode works. - For shortcuts use `key` with `+`-joined names: - `cmd+s` save - `cmd+t` new tab - `cmd+... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-macos-computer-use.md | main | hermes-agent | [
-0.09149360656738281,
-0.005957490298897028,
-0.04301378130912781,
0.07428552955389023,
-0.04176090657711029,
0.0024280836805701256,
0.07681747525930405,
0.030515866354107857,
0.007782995700836182,
-0.021675102412700653,
0.059392981231212616,
-0.07415014505386353,
0.020115483552217484,
-0.... | 0.135218 |
games, anything non-web). - File edits β use `read\_file` / `write\_file` / `patch`, not `type` into an editor window. - Shell commands β use `terminal`, not `type` into Terminal.app. | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-macos-computer-use.md | main | hermes-agent | [
0.032709863036870956,
0.010716953314840794,
-0.03590644150972366,
-0.02750708907842636,
-0.013738707639276981,
0.004573333077132702,
-0.010382738895714283,
0.10926740616559982,
-0.029456852003932,
0.1641591191291809,
-0.014684335328638554,
0.014969931915402412,
-0.03360283002257347,
-0.038... | 0.124264 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Imessage Send and receive iMessages/SMS via the imsg CLI on macOS. ## Skill metadata | | | |---|---| | Source | Bundled (installed by default) | | Path | `skills/apple/im... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/apple/apple-imessage.md | main | hermes-agent | [
-0.07892416417598724,
0.010881636291742325,
0.0150908837094903,
-0.020657572895288467,
-0.01371727418154478,
-0.10097403079271317,
0.05463646724820137,
-0.03588481992483139,
-0.0024565737694501877,
0.016763189807534218,
0.05709821730852127,
0.019255472347140312,
0.07582710683345795,
0.0166... | 0.162777 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Llm Wiki Karpathy's LLM Wiki: build/query interlinked markdown KB. ## Skill metadata | | | |---|---| | Source | Bundled (installed by default) | | Path | `skills/research... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-llm-wiki.md | main | hermes-agent | [
-0.08067447692155838,
0.006638963706791401,
-0.040955930948257446,
0.02235608734190464,
0.018657634034752846,
-0.03291871026158333,
0.026197362691164017,
-0.010941554792225361,
-0.03754957392811775,
0.0392446331679821,
0.017386410385370255,
0.061339009553194046,
0.06385228782892227,
0.0040... | 0.142761 |
that already exist - Missing cross-references to existing content - Contradicting the schema's conventions - Repeating work already logged For large wikis (100+ pages), also run a quick `search\_files` for the topic at hand before creating anything new. ## Initializing a New Wiki When the user asks to create or start a... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-llm-wiki.md | main | hermes-agent | [
-0.022572098299860954,
-0.04359760880470276,
-0.02518007718026638,
0.003949238918721676,
0.05585325136780739,
-0.09784878045320511,
-0.01859905757009983,
0.03780905902385712,
0.00919704232364893,
0.08213106542825699,
0.04551738128066063,
0.01180631946772337,
0.07413652539253235,
0.05381414... | 0.068203 |
passing mentions, minor details, or things outside the domain - \*\*Split a page\*\* when it exceeds ~200 lines β break into sub-topics with cross-links - \*\*Archive a page\*\* when its content is fully superseded β move to `\_archive/`, remove from index ## Entity Pages One page per notable entity. Include: - Overvie... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-llm-wiki.md | main | hermes-agent | [
-0.027230484411120415,
0.0359240397810936,
0.0031378825660794973,
-0.016191475093364716,
0.13475295901298523,
-0.00370810367166996,
-0.03137495741248131,
-0.01948271133005619,
-0.04862960800528526,
0.07015825808048248,
-0.013280225917696953,
0.03462256118655205,
0.041496165096759796,
-0.02... | 0.027093 |
page must link to at least 2 other pages via `[[wikilinks]]`. Check that existing pages link back. - \*\*Tags:\*\* Only use tags from the taxonomy in SCHEMA.md - \*\*Provenance:\*\* On pages synthesizing 3+ sources, append `^[raw/articles/source.md]` markers to paragraphs whose claims trace to a specific source. - \*\*... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-llm-wiki.md | main | hermes-agent | [
-0.0846029669046402,
-0.005314138252288103,
0.0001700006250757724,
0.012346417643129826,
0.13832321763038635,
-0.031740009784698486,
-0.012137715704739094,
-0.0300240870565176,
-0.025291807949543,
0.11630547046661377,
0.01571379415690899,
-0.018930161371827126,
0.10352246463298798,
-0.0076... | 0.051654 |
\*\*Tag audit:\*\* List all tags in use, flag any not in the SCHEMA.md taxonomy. βͺ \*\*Log rotation:\*\* If log.md exceeds 500 entries, rotate it. β« \*\*Report findings\*\* with specific file paths and suggested actions, grouped by severity (broken links > orphans > source drift > contested pages > stale content > styl... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-llm-wiki.md | main | hermes-agent | [
-0.03213607519865036,
0.038786496967077255,
0.02358699031174183,
-0.0004900623462162912,
0.16152577102184296,
-0.06621457636356354,
0.03088429383933544,
-0.03585321083664894,
0.028852539137005806,
0.1028839573264122,
0.052140649408102036,
0.033493947237730026,
0.08235137164592743,
0.048102... | 0.042116 |
and missed cross-references. - \*\*Always update index.md and log.md\*\* β skipping this makes the wiki degrade. These are the navigational backbone. - \*\*Don't create pages for passing mentions\*\* β follow the Page Thresholds in SCHEMA.md. A name appearing once in a footnote doesn't warrant an entity page. - \*\*Don... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-llm-wiki.md | main | hermes-agent | [
-0.11364947259426117,
-0.04693707451224327,
0.03911013901233673,
-0.03570731356739998,
0.0840698778629303,
-0.0787203460931778,
-0.041539326310157776,
-0.008580704219639301,
-0.007215848192572594,
0.06988940387964249,
0.006442980375140905,
0.008212361484766006,
0.03906133398413658,
0.01434... | 0.107557 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Arxiv Search arXiv papers by keyword, author, category, or ID. ## Skill metadata | | | |---|---| | Source | Bundled (installed by default) | | Path | `skills/research/arx... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-arxiv.md | main | hermes-agent | [
-0.09014806896448135,
0.02978789247572422,
-0.06431987881660461,
0.031048400327563286,
-0.006224627140909433,
-0.07111558318138123,
0.029184097424149513,
-0.07675955444574356,
-0.045146506279706955,
0.019593195989727974,
0.02588699385523796,
0.060707759112119675,
0.059469398111104965,
0.00... | 0.118578 |
print(f' title = {{{title}}},') print(f' author = {{{authors}}},') print(f' year = {{{year}}},') print(f' eprint = {{{raw\_id}}},') print(f' archivePrefix = {{arXiv}},') print(f' primaryClass = {{{primary}}},') print(f' url = {{https://arxiv.org/abs/{raw\_id}}}') print('}') " ``` {% endraw %} ## Reading Paper Content A... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-arxiv.md | main | hermes-agent | [
-0.10826936364173889,
0.14197419583797455,
-0.06303051859140396,
0.04475646838545799,
0.05385911464691162,
-0.05227447673678398,
-0.047319523990154266,
-0.06006462872028351,
-0.044292423874139786,
-0.02238890528678894,
0.030359867960214615,
0.009611764922738075,
0.0383676216006279,
-0.0540... | 0.06766 |
to prevent citation drift (a later version may substantially change content) - The API `` field returns the versioned URL (e.g., `http://arxiv.org/abs/1706.03762v7`) ## Withdrawn Papers Papers can be withdrawn after submission. When this happens: - The `` field contains a withdrawal notice (look for "withdrawn" or "ret... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-arxiv.md | main | hermes-agent | [
-0.07603036612272263,
0.06975782662630081,
-0.023664213716983795,
0.028251245617866516,
0.07332092523574829,
-0.04965381324291229,
-0.039360810071229935,
-0.07419207692146301,
0.12719744443893433,
0.034472350031137466,
0.028271911665797234,
0.10595034062862396,
-0.0010984014952555299,
-0.0... | 0.112995 |
{/\* This page is auto-generated from the skill's SKILL.md by website/scripts/generate-skill-docs.py. Edit the source SKILL.md, not this page. \*/} # Research Paper Writing Write ML papers for NeurIPS/ICML/ICLR: designβsubmit. ## Skill metadata | | | |---|---| | Source | Bundled (installed by default) | | Path | `skill... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/skills/bundled/research/research-research-paper-writing.md | main | hermes-agent | [
-0.09861957281827927,
0.005076034925878048,
-0.049495529383420944,
0.020085424184799194,
-0.007379413582384586,
-0.09934095293283463,
-0.02632925473153591,
0.03303854912519455,
-0.04656655341386795,
0.04042782261967659,
0.03250698745250702,
0.008638751693069935,
0.08371926099061966,
-0.037... | 0.204782 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.