content large_stringlengths 3 20.5k | url large_stringlengths 53 192 ⌀ | branch large_stringclasses 4
values | source large_stringclasses 51
values | embeddings listlengths 384 384 | score float64 -0.21 0.65 |
|---|---|---|---|---|---|
mount something via `docker\_volumes`. Security tradeoff: - `false` preserves the sandbox boundary - `true` gives the sandbox direct access to the directory you launched Hermes from Use the opt-in only when you intentionally want the container to work on live host files. ### Persistent Shell By default, each terminal c... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
0.0484711155295372,
0.058577414602041245,
-0.02704550139605999,
-0.015606208704411983,
0.0369291678071022,
-0.06978411972522736,
-0.046672433614730835,
0.012356532737612724,
0.050721291452646255,
0.047100845724344254,
0.033473409712314606,
-0.005175857804715633,
0.034218400716781616,
0.040... | 0.071244 |
This prevents a single read of a minified JS bundle or large data file from flooding the context window. ```yaml file\_read\_max\_chars: 100000 # default — ~25-35K tokens ``` Raise it if you're on a model with a large context window and frequently read big files. Lower it for small-context models to keep reads efficien... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.03045339323580265,
-0.011701149865984917,
0.01794484071433544,
0.009454822167754173,
0.029513508081436157,
-0.09312332421541214,
0.009616688825190067,
0.07393453270196915,
0.07859747856855392,
0.03320881724357605,
0.05374842509627342,
0.08078286796808243,
0.008248916827142239,
-0.038625... | 0.134494 |
call — you can point it at any provider or endpoint. All compression settings live in `config.yaml` (no environment variables). ### Full reference ```yaml compression: enabled: true # Toggle compression on/off threshold: 0.50 # Compress at this % of context limit target\_ratio: 0.20 # Fraction of threshold to preserve ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.049613937735557556,
0.03149695694446564,
-0.012288364581763744,
-0.005293040536344051,
0.0736861601471901,
-0.03446168452501297,
-0.0056054601445794106,
-0.024133753031492233,
0.07985628396272659,
-0.04084368422627449,
-0.009296339005231857,
0.0395621694624424,
-0.044498782604932785,
-0... | 0.136001 |
model — if that model's context window is smaller than the main model's, the summarization call will fail with a context length error. When this happens, the middle turns are \*\*dropped without a summary\*\*, losing conversation context silently. If you override the model, verify its context length meets or exceeds yo... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.05135060101747513,
-0.002093859948217869,
0.027598701417446136,
0.03034103661775589,
0.008747337386012077,
0.011623632162809372,
-0.0224745012819767,
0.03497904911637306,
0.08536819368600845,
-0.02769751287996769,
-0.033390846103429794,
-0.04917372390627861,
0.004369519650936127,
0.0374... | 0.120415 |
httpx waits for the next chunk of data from the provider. Local LLMs can take minutes for prefill on large contexts before producing the first token, so Hermes raises this to 30 minutes when it detects a local endpoint. If you explicitly set `HERMES\_STREAM\_READ\_TIMEOUT`, that value is always used regardless of endpo... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.024676945060491562,
0.011528225615620613,
0.01343762967735529,
-0.006479443982243538,
0.02665351890027523,
-0.07292928546667099,
-0.04147844761610031,
-0.042083755135536194,
0.10253311693668365,
0.01400055456906557,
-0.01989131048321724,
0.01769169420003891,
-0.004696852993220091,
-0.00... | 0.103288 |
count. ## Auxiliary Models Hermes uses "auxiliary" models for side tasks like image analysis, web page summarization, browser screenshot analysis, session-title generation, and context compression. By default (`auxiliary.\*.provider: "auto"`), Hermes routes every auxiliary task to your \*\*main chat model\*\* — the sam... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.04274231940507889,
-0.06124584749341011,
0.013730349950492382,
0.023616477847099304,
0.007182876113802195,
-0.01565050520002842,
-0.04995397478342056,
0.06031399965286255,
0.04592134431004524,
0.02090061642229557,
-0.011908059008419514,
0.01948964037001133,
0.015347159467637539,
0.03354... | 0.161805 |
Hermes is on a remote host see [OAuth over SSH / Remote Hosts](../guides/oauth-over-ssh.md). ::: :::warning `"main"` is for auxiliary tasks only The `"main"` provider option means "use whatever provider my main agent uses" — it's only valid inside `auxiliary:`, `compression:`, and primary fallback entries (`fallback\_p... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
0.04555358365178108,
-0.007257216610014439,
-0.030495421960949898,
0.007388962432742119,
0.05860918015241623,
-0.09160881489515305,
-0.05487377569079399,
0.002345610875636339,
0.06398804485797882,
-0.013812851160764694,
0.0060951365157961845,
-0.02113313414156437,
0.0661487728357315,
0.062... | 0.073817 |
OpenRouter Pareto Code router knob - id: pareto-router min\_coding\_score: 0.5 # 0.0–1.0; higher = stronger coders ``` The shape mirrors what OpenRouter accepts in the chat completions request body. Hermes forwards the entire `extra\_body` verbatim, so any other OpenRouter request-body field documented at [openrouter.a... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.01834338717162609,
-0.018736353144049644,
-0.030286382883787155,
0.02901618927717209,
0.08390218764543533,
-0.14233528077602386,
-0.03359466418623924,
0.03810100257396698,
-0.012335848063230515,
-0.004114598035812378,
-0.024621563032269478,
0.03982430323958397,
-0.06410763412714005,
-0.... | 0.045286 |
"main"` uses whatever provider Hermes uses for normal chat — whether that's a named custom provider (e.g. `beans`), a built-in provider like `openrouter`, or a legacy `OPENAI\_BASE\_URL` endpoint. :::tip If you use Codex OAuth as your main model provider, vision works automatically — no extra configuration needed. Code... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
0.02439015358686447,
-0.021814826875925064,
-0.014884067699313164,
-0.039313480257987976,
0.1059662252664566,
-0.12002330273389816,
-0.03415266424417496,
0.03675707429647446,
0.03491637483239174,
0.007098722737282515,
0.004158765077590942,
-0.0572989322245121,
0.00777078652754426,
0.086078... | 0.072302 |
3. \*\*Google operational guidance\*\* (Gemini and Gemma models only) — conciseness, absolute paths, parallel tool calls, and verify-before-edit patterns. These are transparent to the user and only affect the system prompt. Models that already use tools reliably (like Claude) don't need this guidance, which is why `"au... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.05798996612429619,
-0.029010586440563202,
0.03514096885919571,
-0.025729738175868988,
-0.018242808058857918,
-0.01243655476719141,
0.08034183830022812,
0.014928403310477734,
-0.027148757129907608,
-0.015947945415973663,
0.019488373771309853,
-0.041334379464387894,
0.008878077380359173,
... | 0.083511 |
fr | tr | uk | af | ko | it | ga | pt | ru | hu ``` ### File-mutation verifier When `display.file\_mutation\_verifier` is `true` (default), Hermes appends a one-line advisory to the assistant's final response whenever a `write\_file` or `patch` call failed during the turn and was never superseded by a successful write ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.13023772835731506,
-0.0008379233186133206,
0.028476377949118614,
-0.013790399767458439,
0.016368582844734192,
-0.09476909786462784,
-0.04914117977023125,
0.056111179292201996,
-0.00941384956240654,
0.08720462769269943,
0.029337584972381592,
0.020769037306308746,
0.04390282183885574,
-0.... | 0.092076 |
noisy. Use `display.platforms` to set per-platform modes: ```yaml display: tool\_progress: all # global default platforms: signal: tool\_progress: 'off' # silence progress on Signal telegram: tool\_progress: verbose # detailed progress on Telegram slack: tool\_progress: 'off' # quiet in shared Slack workspace ``` Platf... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.015371095389127731,
-0.06211123988032341,
0.0363745242357254,
-0.0024897190742194653,
-0.01152994018048048,
-0.09941080212593079,
0.032197944819927216,
-0.04521460086107254,
0.010114521719515324,
0.023951342329382896,
0.03236962482333183,
-0.017301145941019058,
0.005320227704942226,
0.0... | 0.122081 |
```yaml streaming: enabled: true # Enable progressive message editing transport: edit # "edit" (progressive message editing) or "off" edit\_interval: 0.3 # Seconds between message edits buffer\_threshold: 40 # Characters before forcing an edit flush cursor: " ▉" # Cursor shown during streaming fresh\_final\_after\_seco... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.034419212490320206,
-0.05710794776678085,
0.09122499078512192,
0.009034780785441399,
0.0188919547945261,
-0.03409038484096527,
-0.043340716511011124,
0.00010999140795320272,
0.0813683569431305,
0.048078589141368866,
0.014775551855564117,
-0.004092847462743521,
-0.029792308807373047,
0.0... | 0.150646 |
run locally on the host and return the output directly — no LLM call, no tokens consumed. `alias` commands rewrite to the configured slash command target. - \*\*30-second timeout\*\* — long-running commands are killed with an error message - \*\*Priority\*\* — quick commands are checked before skill commands, so you ca... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.021026523783802986,
-0.03256847336888313,
0.04196159914135933,
0.04186791181564331,
-0.011535274796187878,
-0.015236028470098972,
0.03211141377687454,
0.009364364668726921,
0.022565238177776337,
0.058513056486845016,
0.02767937071621418,
-0.10479209572076797,
-0.005825497675687075,
0.04... | 0.091463 |
Supports `category` filtering (`company`, `research paper`, `news`, `people`, `personal site`, `pdf`) and domain/date filters. ## Browser Configure browser automation behavior: ```yaml browser: inactivity\_timeout: 120 # Seconds before auto-closing idle sessions command\_timeout: 30 # Timeout in seconds for browser com... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.0189075730741024,
0.011873801238834858,
-0.023663854226469994,
-0.013511423021554947,
0.047373346984386444,
-0.05628911033272743,
-0.034754782915115356,
-0.023975064978003502,
-0.02540259249508381,
-0.07795543223619461,
-0.007640117313712835,
-0.008960873819887638,
-0.022194957360625267,
... | 0.050315 |
the conversation context and logs. \*\*On by default\*\*. Set to `false` explicitly only when you need raw credential-like strings for debugging or redactor development. - `tirith\_enabled` — when `true`, terminal commands are scanned by [Tirith](https://github.com/sheeki03/tirith) before execution to detect potentiall... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.03838745504617691,
-0.01788034848868847,
-0.07886522263288498,
-0.00016073677397798747,
-0.008720831014215946,
-0.11760412156581879,
0.029004806652665138,
-0.05850145220756531,
0.0074675120413303375,
0.006757775787264109,
0.04048077389597893,
-0.02052430994808674,
0.046361807733774185,
... | 0.166207 |
(default): parent spawns leaves that cannot delegate. 2 = orchestrator children can spawn leaf grandchildren. 3 = three levels. orchestrator\_enabled: true # Global kill switch. When false, role="orchestrator" is ignored and every child is forced to leaf regardless of max\_spawn\_depth. ``` \*\*Subagent provider:model ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.019839299842715263,
-0.029489602893590927,
-0.00943569652736187,
0.01846250146627426,
0.016546033322811127,
-0.050521016120910645,
-0.02695426158607006,
0.009813645854592323,
0.03396082669496536,
0.02066308818757534,
-0.014095064252614975,
-0.08391344547271729,
0.05764208361506462,
0.04... | 0.075356 |
Working directory only | - \*\*SOUL.md\*\* is the agent's primary identity. It occupies slot #1 in the system prompt, completely replacing the built-in default identity. Edit it to fully customize who the agent is. - If SOUL.md is missing, empty, or cannot be loaded, Hermes falls back to a built-in default identity. - ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/configuration.md | main | hermes-agent | [
-0.02555094100534916,
0.038878295570611954,
-0.048110537230968475,
-0.06512381881475449,
0.0040757544338703156,
-0.05919240787625313,
0.047247543931007385,
-0.020441705361008644,
0.03996077552437782,
0.018734151497483253,
0.027765687555074692,
-0.005250200629234314,
0.05719531327486038,
-0... | 0.098762 |
import useBaseUrl from '@docusaurus/useBaseUrl'; # Sessions Hermes Agent automatically saves every conversation as a session. Sessions enable conversation resume, cross-session search, and full conversation history management. ## How Sessions Work Every conversation — whether from the CLI, Telegram, Discord, Slack, Wha... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/sessions.md | main | hermes-agent | [
-0.05433114990592003,
-0.017587726935744286,
-0.027250193059444427,
-0.008922802284359932,
0.018559983000159264,
-0.05258045718073845,
0.03197863698005676,
0.010114437900483608,
0.05471639707684517,
0.05098439380526543,
-0.05092230066657066,
0.00946011021733284,
-0.0033488941844552755,
-0.... | 0.130745 |
| | `feishu` | Feishu/Lark messenger | | `wecom` | WeCom (WeChat Work) | | `weixin` | Weixin (personal WeChat) | | `bluebubbles` | Apple iMessage via BlueBubbles macOS server | | `qqbot` | QQ Bot (Tencent QQ) via Official API v2 | | `homeassistant` | Home Assistant conversation | | `webhook` | Incoming webhooks | | `ap... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/sessions.md | main | hermes-agent | [
-0.10617014020681381,
-0.02701704576611519,
-0.02453312836587429,
-0.0015647955005988479,
-0.029661037027835846,
-0.030503302812576294,
0.0278953704982996,
0.019172806292772293,
-0.0032151544000953436,
-0.016659758985042572,
0.020143140107393265,
-0.03141335770487785,
0.006026274990290403,
... | 0.171994 |
— wait for the current response to finish first. 3. The gateway watcher claims the handoff and asks the destination adapter for a fresh thread: - \*\*Telegram\*\* — opens a new forum topic (DM topics if Bot API 9.4+ Topics mode is enabled in the chat, or a forum supergroup topic). - \*\*Discord\*\* — creates a 1440-min... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/sessions.md | main | hermes-agent | [
-0.08569364994764328,
-0.05692313611507416,
0.03505398705601692,
-0.02913070097565651,
0.003530822228640318,
-0.019865525886416435,
0.052604757249355316,
-0.03097984939813614,
0.0923045426607132,
0.045864831656217575,
0.016941286623477936,
0.013034753501415253,
-0.036095164716243744,
0.035... | 0.090277 |
same title - \*\*Max 100 characters\*\* — keeps listing output clean - \*\*Sanitized\*\* — control characters, zero-width chars, and RTL overrides are stripped automatically - \*\*Normal Unicode is fine\*\* — emoji, CJK, accented characters all work ### Auto-Lineage on Compression When a session's context is compressed... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/sessions.md | main | hermes-agent | [
-0.10305323451757431,
0.0527445413172245,
0.06803010404109955,
-0.028009701520204544,
-0.04667096212506294,
-0.053478579968214035,
0.04239644110202789,
-0.015234866179525852,
0.07973992824554443,
0.04110503941774368,
-0.05005320906639099,
-0.01592620648443699,
0.02446596883237362,
0.042759... | 0.078739 |
and lets the agent scroll through any session it finds. No LLM calls, no summarization, no truncation. Every shape returns actual messages from the DB. ### Three calling shapes The tool infers what you want from which arguments you set. There's no `mode` parameter. \*\*1. Discovery — pass `query`:\*\* ```python session... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/sessions.md | main | hermes-agent | [
-0.03164759650826454,
0.021636033430695534,
-0.07022105157375336,
0.03689465671777725,
-0.002544558374211192,
-0.04871376231312752,
0.04595610126852989,
0.0019582132808864117,
0.006010499317198992,
-0.01783996820449829,
-0.0588078536093235,
-0.010845228098332882,
0.05351140350103378,
0.011... | 0.095589 |
| | Group thread/topic | `agent:main:<platform>:group:<chat\_id>:<thread\_id>` | Shared session for all thread participants (default). Per-user with `thread\_sessions\_per\_user: true`. | | Channel | `agent:main:<platform>:channel:<chat\_id>:<user\_id>` | Per-user inside the channel when the platform exposes a user ID ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/sessions.md | main | hermes-agent | [
-0.009066264145076275,
-0.006986163556575775,
-0.04303918406367302,
-0.03281720355153084,
0.0012371604098007083,
-0.0394669808447361,
0.06755784153938293,
-0.06564803421497345,
0.06844941526651382,
0.021619072183966637,
-0.022432923316955566,
-0.03081962838768959,
0.019345056265592575,
0.0... | 0.118747 |
sessions this many days vacuum\_after\_prune: true # reclaim disk space after a pruning sweep min\_interval\_hours: 24 # don't re-run the sweep more often than this ``` Active sessions are never auto-pruned, regardless of age. ### Manual Cleanup ```bash # Prune sessions older than 90 days hermes sessions prune # Delete... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/sessions.md | main | hermes-agent | [
-0.04239315167069435,
0.05367572233080864,
-0.03580145165324211,
0.07201801240444183,
0.043143581598997116,
-0.06086508557200432,
0.011485854163765907,
-0.002361765829846263,
-0.02420402318239212,
0.051347095519304276,
0.001933449413627386,
0.10854262858629227,
0.0037669572047889233,
0.028... | 0.094839 |
# Checkpoints and `/rollback` Hermes Agent can automatically snapshot your project before \*\*destructive operations\*\* and restore it with a single command. Checkpoints are \*\*opt-in\*\* as of v2 — most users never use `/rollback`, and the shadow-store storage is non-trivial over time, so the default is off. Enable ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/checkpoints-and-rollback.md | main | hermes-agent | [
-0.05109746754169464,
-0.0017203678144142032,
-0.03952904790639877,
0.00602307403460145,
0.025655271485447884,
-0.1032818928360939,
0.01669916696846485,
-0.05131691321730614,
0.07149925827980042,
0.08060586452484131,
0.05884043499827385,
0.07784971594810486,
0.04599345102906227,
-0.0555918... | 0.073695 |
false auto\_prune: false ``` When `enabled: false`, the Checkpoint Manager is a no-op and never attempts git operations. When `auto\_prune: false`, the store grows until you run `hermes checkpoints prune` manually. ## Listing Checkpoints From a CLI session: ``` /rollback ``` Hermes responds with a formatted list showin... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/checkpoints-and-rollback.md | main | hermes-agent | [
-0.021874716505408287,
0.05168086662888527,
-0.03009570762515068,
-0.031121090054512024,
0.05673452094197273,
-0.105878084897995,
0.026466993615031242,
0.014478436671197414,
0.013809899799525738,
0.060092467814683914,
0.11769179999828339,
0.08956858515739441,
0.04067707806825638,
-0.052129... | 0.002648 |
`prune` / `clear` instead. ### Migration from v1 Before the v2 rewrite, each working directory got its own complete shadow git repo directly under `~/.hermes/checkpoints//`. That layout couldn't dedup objects across projects and had a documented no-op pruner — the store would grow without bound. On first v2 run, any pr... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/checkpoints-and-rollback.md | main | hermes-agent | [
-0.04071813449263573,
0.061534784734249115,
0.013305353932082653,
-0.041655708104372025,
0.020016813650727272,
-0.08539668470621109,
-0.09721101820468903,
-0.05368372052907944,
-0.019967660307884216,
0.028521627187728882,
0.09100934863090515,
0.09562904387712479,
0.026646921411156654,
-0.0... | -0.061264 |
# Hermes Agent — Docker There are two distinct ways Docker intersects with Hermes Agent: 1. \*\*Running Hermes IN Docker\*\* — the agent itself runs inside a container (this page's primary focus) 2. \*\*Docker as a terminal backend\*\* — the agent runs on your host but executes every command inside a single, persistent... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
-0.018023598939180374,
0.06880700588226318,
-0.0392301082611084,
-0.03665180504322052,
0.02696785144507885,
-0.12172374874353409,
-0.01875441148877144,
-0.017846979200839996,
0.00427645817399025,
0.050490520894527435,
0.004263428505510092,
-0.004628851544111967,
0.0955343022942543,
-0.0420... | 0.091231 |
\ --restart unless-stopped \ -v ~/.hermes:/opt/data \ -p 8642:8642 \ -e API\_SERVER\_ENABLED=true \ -e API\_SERVER\_HOST=0.0.0.0 \ -e API\_SERVER\_KEY="$(openssl rand -hex 32)" \ -e API\_SERVER\_CORS\_ORIGINS='\*' \ nousresearch/hermes-agent gateway run ``` Opening any port on an internet facing machine is a security r... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
-0.027241356670856476,
0.04623224586248398,
-0.09734097123146057,
-0.06746653467416763,
0.01759839989244938,
-0.055491454899311066,
-0.06356196105480194,
-0.0016503653023391962,
0.00034581543877720833,
0.0036995327100157738,
-0.0005716168088838458,
-0.006344190798699856,
0.05940302088856697,... | 0.039467 |
process. ## Running interactively (CLI chat) To open an interactive chat session against a running data directory: ```sh docker run -it --rm \ -v ~/.hermes:/opt/data \ nousresearch/hermes-agent ``` Or if you have already opened a terminal in your running container (via Docker Desktop for instance), just run: ```sh /opt... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
-0.0252927727997303,
0.030617522075772285,
-0.05463145673274994,
-0.04093009978532791,
0.0046864766627550125,
-0.10455335676670074,
-0.013171946629881859,
0.006576335057616234,
0.05847106873989105,
0.027122680097818375,
-0.01954449899494648,
-0.03132733330130577,
0.045759763568639755,
-0.0... | 0.083134 |
### Why one container with many profiles, not many containers Before the s6 migration, "one container per profile" was the recommended pattern because there was no in-container supervisor to manage multiple gateways. With s6 as PID 1, that's no longer necessary, and the single-container layout is simpler in almost ever... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
0.023059412837028503,
-0.015869662165641785,
-0.014931614510715008,
-0.031635183840990067,
0.04388919472694397,
-0.006517794914543629,
-0.0758635625243187,
0.03693830966949463,
-0.047878947108983994,
0.013940932229161263,
-0.01508378330618143,
0.03360772132873535,
0.06791673600673676,
-0.0... | 0.047415 |
what survives container restarts. `docker logs` only retains output from the current container's lifetime (and is wiped on `docker rm`); the rotated files persist on the bind-mounted volume. - The boot reconciler's audit line shape is ` profile= prior\_state= action=`, so a quick `grep profile=coder ~/.hermes/logs/cont... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
-0.015883196145296097,
0.06003532186150551,
-0.016274280846118927,
-0.012997708283364773,
0.06351808458566666,
-0.022398479282855988,
-0.049501724541187286,
-0.028616806492209435,
0.0294659286737442,
0.01786554977297783,
0.0010810969397425652,
0.03536771237850189,
0.033412784337997437,
-0.... | 0.027208 |
Start it with your host UID/GID so the container process can access the per-user audio socket: ```sh export HERMES\_UID="$(id -u)" export HERMES\_GID="$(id -g)" docker compose up -d --build ``` To verify what PortAudio sees inside the container: ```sh docker exec hermes /opt/hermes/.venv/bin/python -c "import sounddevi... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
0.05649805814027786,
-0.01810918189585209,
-0.031057506799697876,
-0.04202817380428314,
-0.04584043100476265,
-0.06195394694805145,
-0.055163826793432236,
-0.004850912373512983,
-0.06947743147611618,
0.030643409118056297,
-0.040295880287885666,
-0.06449539959430695,
0.00505380192771554,
-0... | 0.067528 |
the main program. Starting `hermes gateway run` as root inside the official image is refused by default because it can leave root-owned files in `/opt/data` and break later dashboard or gateway starts. Set `HERMES\_ALLOW\_ROOT\_GATEWAY=1` only when you intentionally accept that risk. ::: ### `docker exec` automatically... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
0.0033545976039022207,
0.030825331807136536,
-0.0414719395339489,
-0.028806399554014206,
0.026615353301167488,
-0.1336461454629898,
-0.04899666830897331,
0.024985622614622116,
0.030577486380934715,
0.06069327890872955,
0.012126616202294827,
-0.010450483299791813,
0.050038352608680725,
0.00... | 0.066747 |
container as read-only volumes. Skill scripts, templates, and references are available inside the sandbox without manual configuration, and because the container persists for the life of the Hermes process, any dependencies you install or files you write stay around for the next tool call. The same syncing happens for ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
0.009492523968219757,
0.025994932278990746,
-0.018485717475414276,
-0.015398452989757061,
0.0031888652592897415,
-0.053303588181734085,
-0.06706036627292633,
0.010353081859648228,
-0.0027695850003510714,
0.02632725052535534,
-0.010090162046253681,
0.03762676194310188,
0.03967281058430672,
... | 0.072544 |
networks: hermes-net: driver: bridge ``` From inside the Hermes container, the sidecar is reachable at `http://my-tool:` (or whatever protocol it serves). This pattern keeps each service's lifecycle, resource limits, and upgrade cadence independent, and avoids bloating the Hermes image with dependencies that are only n... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
-0.0541955828666687,
0.06995529681444168,
0.012167627923190594,
-0.028898172080516815,
0.038879118859767914,
-0.09933066368103027,
-0.05725541710853577,
0.006606890819966793,
-0.005977765657007694,
0.028084339573979378,
0.0053943139500916,
-0.001731158117763698,
0.02759629301726818,
-0.038... | 0.109053 |
Common causes: - Missing or invalid `.env` file — run interactively first to complete setup - Port conflicts if running with exposed ports ### "Permission denied" errors The container's stage2 hook drops privileges to the non-root `hermes` user (UID 10000) via `s6-setuidgid` inside each supervised service. If your host... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/docker.md | main | hermes-agent | [
0.010073160752654076,
-0.04933857172727585,
-0.01707015186548233,
-0.0398099385201931,
-0.001340836868621409,
-0.14385443925857544,
-0.06559642404317856,
-0.005974076222628355,
-0.0480986163020134,
0.03312879055738449,
0.03527483344078064,
-0.03756830468773842,
0.025414621457457542,
0.0450... | 0.029449 |
# Git Worktrees Hermes Agent is often used on large, long‑lived repositories. When you want to: - Run \*\*multiple agents in parallel\*\* on the same project, or - Keep experimental refactors isolated from your main branch, Git \*\*worktrees\*\* are the safest way to give each agent its own checkout without duplicating... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/git-worktrees.md | main | hermes-agent | [
-0.028600720688700676,
-0.012996109202504158,
-0.021597040817141533,
-0.03715555742383003,
0.0006209269049577415,
-0.1483704000711441,
0.009399527683854103,
-0.02418764866888523,
0.1053292378783226,
0.07071585953235626,
0.027154892683029175,
0.03420703113079071,
0.0413905568420887,
-0.0079... | 0.103821 |
after the experiment\*\* - e.g. `feature/hermes-checkpoints-docs`, `feature/hermes-refactor-tests`. - \*\*Commit frequently\*\* - Use git commits for high‑level milestones. - Use [checkpoints and /rollback](./checkpoints-and-rollback.md) as a safety net for tool‑driven edits in between. - \*\*Avoid running Hermes from ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/git-worktrees.md | main | hermes-agent | [
-0.04818274825811386,
0.02294832468032837,
0.003637379966676235,
0.011552263982594013,
-0.011805576272308826,
-0.10137543827295303,
-0.06464722007513046,
-0.02404213324189186,
0.025865141302347183,
0.08054729551076889,
0.005000154487788677,
0.0018723215907812119,
0.06342893838882446,
-0.00... | 0.040919 |
# Security Hermes Agent is designed with a defense-in-depth security model. This page covers every security boundary — from command approval to container isolation to user authorization on messaging platforms. ## Overview The security model has seven layers: 1. \*\*User authorization\*\* — who can talk to the agent (al... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
-0.03473341464996338,
0.04591662809252739,
-0.043456047773361206,
-0.07937043905258179,
0.02921900525689125,
-0.06572885811328888,
0.041162800043821335,
0.030397551134228706,
-0.008809677325189114,
0.03291766345500946,
-0.006351548712700605,
-0.038001298904418945,
0.1115933507680893,
0.000... | 0.180696 |
--yolo` 2. \*\*Slash command\*\*: Type `/yolo` during a session to toggle it on/off 3. \*\*Environment variable\*\*: Set `HERMES\_YOLO\_MODE=1` The `/yolo` command is a \*\*toggle\*\* — each use flips the mode on or off: ``` > /yolo ⚡ YOLO mode ON — all commands auto-approved. Use with caution. > /yolo ⚠ YOLO mode OFF ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
-0.04062746465206146,
0.01908416673541069,
0.05423278734087944,
-0.016795361414551735,
0.03215227276086807,
-0.04789666086435318,
0.06417027860879898,
-0.005355945788323879,
0.016068430617451668,
-0.004875865764915943,
0.03143543377518654,
-0.04758361726999283,
-0.007817840203642845,
-0.00... | 0.196444 |
| `chmod --recursive` with unsafe perms | Recursive world/other-writable (long flag) | | `chown -R root` / `chown --recursive root` | Recursive chown to root | | `mkfs` | Format filesystem | | `dd if=` | Disk copy | | `> /dev/sd` | Write to block device | | `DROP TABLE/DATABASE` | SQL DROP | | `DELETE FROM` (without WH... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
0.0025209039449691772,
0.010346642695367336,
-0.09414511919021606,
-0.07110752165317535,
0.045672740787267685,
-0.0444357730448246,
0.06635450571775436,
0.08024699985980988,
-0.047733452171087265,
0.07740239799022675,
0.09008696675300598,
-0.06083020567893982,
0.029704414308071136,
0.00744... | 0.072373 |
\*\*Platform-specific allowlists\*\* (e.g., `TELEGRAM\_ALLOWED\_USERS=12345,67890`) 4. \*\*Global allowlist\*\* (`GATEWAY\_ALLOWED\_USERS=12345,67890`) 5. \*\*Global allow-all\*\* (`GATEWAY\_ALLOW\_ALL\_USERS=true`) 6. \*\*Default: deny\*\* ### Platform Allowlists Set allowed user IDs as comma-separated values in `~/.h... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
0.02771829254925251,
0.040329378098249435,
-0.03939257934689522,
-0.12379065901041031,
0.015558022074401379,
-0.02335820533335209,
0.10850147157907486,
-0.05426542088389397,
-0.05494987592101097,
0.05640321597456932,
0.03218977898359299,
-0.04215339943766594,
0.04033829644322395,
0.0552612... | 0.088991 |
overlay2 on XFS) container_persistent: true # Persist filesystem across sessions ``` ### Filesystem Persistence - **Persistent mode** (`container_persistent: true`): Bind-mounts `/workspace` and `/root` from `~/.hermes/sandboxes/docker//` - **Ephemeral mode** (`container_persistent: false`): Uses tmpfs for workspace — ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
0.03726154565811157,
0.020885780453681946,
-0.03459208086133003,
-0.017268149182200432,
0.12639360129833221,
-0.11320867389440536,
-0.017033396288752556,
0.051533233374357224,
0.005485024768859148,
0.02075750008225441,
0.03628917410969734,
-0.012684766203165054,
0.06957616657018661,
0.0130... | 0.080556 |
container. This list is read by `tools/credential_files.py` (`terminal.credential_files`) — it lives under the `terminal:` block but is loaded by the credential-files module, not the core terminal backend, so it isn't part of the bundled `DEFAULT_CONFIG` snapshot. ### What Each Sandbox Filters | Sandbox | Default Filte... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
-0.017684275284409523,
0.017831547185778618,
-0.09020812809467316,
-0.06785503029823303,
0.05025666579604149,
-0.04870038107037544,
0.0030985563062131405,
0.05792287364602089,
-0.0542893260717392,
-0.0056213149800896645,
0.030671268701553345,
-0.008408532477915287,
0.07334204018115997,
-0.... | 0.064161 |
DNS failures are treated as blocked (fail-closed). Redirect chains are re-validated at each hop to prevent redirect-based bypasses. #### Intentionally allowing private URLs Some setups legitimately need private/internal URL access — home networks that resolve `home.arpa` to RFC 1918 space, LAN-only Ollama/llama.cpp end... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
0.006902710534632206,
-0.041161779314279556,
0.06719497591257095,
-0.016552293673157692,
-0.015835434198379517,
-0.05766182020306587,
-0.06952694803476334,
-0.06395338475704193,
0.07769775390625,
0.014170345850288868,
0.010521655902266502,
0.07358983904123306,
-0.057402756065130234,
-0.008... | 0.023986 |
never run the gateway as root 9. **Monitor logs** — check `~/.hermes/logs/` for unauthorized access attempts 10. **Keep updated** — run `hermes update` regularly for security patches ### Securing API Keys ```bash # Set proper permissions on the .env file chmod 600 ~/.hermes/.env # Keep separate keys for different servi... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
0.05889467895030975,
0.028268348425626755,
-0.0853002592921257,
-0.03718157857656479,
-0.00963976327329874,
-0.052172210067510605,
-0.07975704222917557,
-0.010288898833096027,
-0.012548347935080528,
0.04701239988207817,
-0.0011261010076850653,
0.004325934685766697,
0.04012608528137207,
-0.... | -0.004443 |
— a malicious `config.yaml` cannot redirect the install | | Allowlist | Only specs that appear in the in-tree `LAZY_DEPS` map can be installed via this path. A typo in a feature name does NOT get install-anything semantics | | Opt-out | Set `security.allow_lazy_installs: false` to disable runtime installs entirely. Use... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/security.md | main | hermes-agent | [
-0.0012376519152894616,
0.05439314246177673,
0.02138957567512989,
-0.06919308006763458,
0.03703130781650543,
-0.10028403252363205,
-0.030973728746175766,
0.013046368025243282,
-0.1175658255815506,
0.028847869485616684,
0.09367210417985916,
0.031711384654045105,
0.023091431707143784,
0.0082... | 0.051739 |
# Bitwarden Secrets Manager Pull API keys from [Bitwarden Secrets Manager](https://bitwarden.com/products/secrets-manager/) at process startup instead of storing them in plaintext inside `~/.hermes/.env`. One bootstrap secret (a machine-account access token) replaces N per-provider keys, and rotating a credential becom... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/secrets/bitwarden.md | main | hermes-agent | [
-0.0921502560377121,
0.0016164304688572884,
-0.0495748445391655,
-0.022835612297058105,
-0.034798361361026764,
-0.07960889488458633,
0.008308564312756062,
0.013305277563631535,
-0.0034057325683534145,
0.004929554648697376,
0.012911063618957996,
-0.031075727194547653,
0.104807049036026,
-0.... | 0.016547 |
the first time secrets are applied in a process. ## CLI | Command | What it does | |---|---| | `hermes secrets bitwarden setup` | Interactive wizard (install binary, prompt for token, pick project, test fetch) | | `hermes secrets bitwarden status` | Show config + binary version + token presence | | `hermes secrets bitw... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/secrets/bitwarden.md | main | hermes-agent | [
-0.06627216935157776,
-0.01892935112118721,
-0.04592183232307434,
0.013598007149994373,
-0.008148367516696453,
-0.05641583725810051,
-0.02321803569793701,
-0.04104215279221535,
-0.011861329898238182,
0.018299046903848648,
0.019443051889538765,
-0.039427340030670166,
0.04788977652788162,
-0... | 0.04671 |
itself, even with `override\_existing: true`. If you store `BWS\_ACCESS\_TOKEN` as a secret inside the project, it's silently skipped during apply. - The `bws` binary download is verified against the published SHA-256 checksum from the same GitHub release. Mismatch aborts the install. - The pinned version (`bws v2.0.0`... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/secrets/bitwarden.md | main | hermes-agent | [
-0.033790383487939835,
-0.03907809033989906,
0.0015775852371007204,
-0.01591815985739231,
0.0033039222471415997,
-0.08221394568681717,
-0.050254419445991516,
-0.0650079995393753,
-0.03252090886235237,
0.019795527681708336,
0.04267352819442749,
-0.0010971557348966599,
0.05855409801006317,
-... | -0.01314 |
# Secrets Hermes can pull API keys from external secret managers at process startup instead of storing them in `~/.hermes/.env`. The bootstrap token for the secret manager lives in `.env`; every other provider key (OpenAI, Anthropic, OpenRouter, etc.) can stay in the manager and rotate centrally. Supported: - [Bitwarde... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/secrets/index.md | main | hermes-agent | [
-0.10187039524316788,
-0.0021293729078024626,
-0.05907914415001869,
0.01604059524834156,
0.014276789501309395,
-0.04700888320803642,
-0.030789371579885483,
0.0050079552456736565,
0.020849382504820824,
0.0434497594833374,
0.01734934374690056,
0.022589145228266716,
0.08437635004520416,
-0.06... | 0.040231 |
# Deliverable Mode When Hermes Agent runs inside a messaging gateway (Slack, Discord, Telegram, WhatsApp, Signal, etc.), it can deliver generated files directly into the chat — not as paths the user has to copy, but as native attachments. A chart shows up as an inline image. A PDF report shows up as a file download. A ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/deliverable-mode.md | main | hermes-agent | [
-0.03176206350326538,
0.01897457055747509,
-0.06575809419155121,
-0.011440534144639969,
0.027774402871727943,
-0.10815779864788055,
-0.04431852698326111,
0.012793032452464104,
0.07480333000421524,
0.019871270284056664,
0.0299282968044281,
0.08668841421604156,
0.0630696564912796,
-0.0034603... | 0.152704 |
completed" message to whoever subscribed to the task in Slack/Telegram/etc., it also uploads each artifact as a native attachment to that chat. The human gets the deliverable and the summary in one place. Files that don't exist on disk when the notifier runs are silently skipped. ## Connecting more services with MCP Be... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/deliverable-mode.md | main | hermes-agent | [
-0.03669067472219467,
-0.052634723484516144,
0.057760488241910934,
0.005232519004493952,
0.010877548716962337,
-0.08135596662759781,
0.009557717479765415,
-0.03387390077114105,
-0.0029712561517953873,
0.09976308792829514,
-0.009158297441899776,
-0.038612835109233856,
0.022140871733427048,
... | 0.231638 |
# Web Dashboard The web dashboard is a browser-based UI for managing your Hermes Agent installation. Instead of editing YAML files or running CLI commands, you can configure settings, manage API keys, and monitor sessions from a clean web interface. :::tip Hosted-mode auth uses Nous Portal OAuth; if you also want the d... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
0.00007056556933093816,
0.031839821487665176,
-0.09318841248750687,
-0.07593396306037903,
0.010626575909554958,
-0.09986425191164017,
-0.05063623934984207,
-0.011387867853045464,
0.0381939597427845,
0.029990179464221,
0.055843535810709,
-0.007524826098233461,
0.030561836436390877,
-0.01646... | 0.074248 |
The server spawns `hermes --tui` behind a POSIX pseudo-terminal - Keystrokes travel to the PTY; ANSI output streams back to the browser - xterm.js's WebGL renderer paints each cell to an integer-pixel grid; mouse tracking (SGR 1006), wide characters (Unicode 11), and box-drawing glyphs all render natively - Resizing th... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.047768495976924896,
-0.030230175703763962,
-0.026317505165934563,
-0.02182331681251526,
-0.004360706079751253,
-0.035577237606048584,
-0.008415981195867062,
-0.022213101387023926,
-0.016188696026802063,
0.009216982871294022,
0.0064001064747571945,
0.013294107280671597,
-0.0264365561306476... | 0.065636 |
— full-text search across all message content using FTS5. Results show highlighted snippets and auto-scroll to the first matching message when expanded. - \*\*Stats\*\* — a summary bar shows total sessions, how many are active in the store, archived count, total messages, and a per-source breakdown. - \*\*Expand\*\* — ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.03379611298441887,
0.057925157248973846,
-0.002254405990242958,
0.11407346278429031,
0.10663822293281555,
-0.006569553632289171,
0.054465603083372116,
-0.007633179891854525,
0.043663762509822845,
0.0033557992428541183,
-0.04389628767967224,
0.03979964181780815,
0.023936130106449127,
0.0... | 0.144305 |
Changes take effect on the next session. - \*\*Toolsets\*\* — a separate view shows built-in toolsets (file operations, web browsing, etc.) with their active/inactive status, setup requirements, and list of included tools - \*\*Browse hub\*\* — a third view searches the skill hub across all sources (the same as `hermes... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.05578102916479111,
-0.057974521070718765,
-0.011103004217147827,
-0.00647024204954505,
0.023668259382247925,
-0.0606720969080925,
-0.029111506417393684,
-0.08057533204555511,
-0.08011042326688766,
0.020172739401459694,
0.05176994949579239,
0.012341811321675777,
0.05703215301036835,
-0.0... | 0.071365 |
validated. A "Setup guide" link points to the platform's credential docs. - \*\*Enable / disable\*\* — toggle a channel on or off. The credential stays on disk; only the active state changes. - \*\*Test\*\* — check whether the channel is configured, enabled, and reporting a live connection from the gateway. - \*\*Resta... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.009639858268201351,
-0.06464904546737671,
-0.09013871103525162,
-0.04672124236822128,
-0.050016481429338455,
-0.04240149259567261,
-0.021174229681491852,
-0.018376143649220467,
-0.005692183505743742,
0.016341013833880424,
-0.0004683054576162249,
0.04941151291131973,
0.048729248344898224,
... | 0.076652 |
its own. ::: ## `/reload` Slash Command The dashboard PR also adds a `/reload` slash command to the interactive CLI. After changing API keys via the web dashboard (or by editing `.env` directly), use `/reload` in an active CLI session to pick up the changes without restarting: ``` You → /reload Reloaded .env (3 var(s) ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.06413524597883224,
0.011775451712310314,
-0.06273435801267624,
-0.03522532805800438,
0.0019171149469912052,
-0.05689242482185364,
-0.022428585216403008,
0.01675371639430523,
0.07042212784290314,
0.053486619144678116,
0.018439892679452896,
0.03761185333132744,
0.055153850466012955,
-0.00... | 0.160281 |
channel with status + per-platform setup fields | | `PUT /api/messaging/platforms/{id}` | Configure a channel. Body: `{enabled?, env?, clear\_env?}` (env writes to `.env`, enabled to `config.yaml`) | | `POST /api/messaging/platforms/{id}/test` | Report whether a channel is configured, enabled, and connected | | `GET /a... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.03844763711094856,
0.014842310920357704,
-0.05439630150794983,
0.004737266805022955,
-0.005639039911329746,
-0.06282699108123779,
0.05058325454592705,
-0.04040371999144554,
0.0022442687768489122,
0.02931436523795128,
0.07251147925853729,
0.023773811757564545,
0.0363432839512825,
0.05669... | 0.066712 |
| The gate is on if and only if: 1. The bind host is not `127.0.0.1`, `::1`, `localhost`, or `0.0.0.0` AND 2. The `--insecure` flag is \*\*not\*\* set. Setting `--insecure` keeps the existing single-process session-token behaviour — no OAuth dance, no provider plugins required. Use only on networks where you trust ever... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.05679025501012802,
0.02971082180738449,
-0.07001867890357971,
-0.050645988434553146,
0.0021747269202023745,
-0.09144865721464157,
-0.027889642864465714,
-0.03870747610926628,
0.03789182007312775,
0.024884158745408058,
0.046790555119514465,
-0.04256325960159302,
0.06237458065152168,
0.05... | 0.05607 |
as the other dashboard settings — env wins over `config.yaml`: | Surface | Override path | When to use | |---------|---------------|-------------| | `dashboard.public\_url` in `config.yaml` | `HERMES\_DASHBOARD\_PUBLIC\_URL` | Local dev / on-prem (canonical) | | `HERMES\_DASHBOARD\_PUBLIC\_URL` env var | — | Fly.io pla... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.04487122222781181,
0.03757590800523758,
0.015424107201397419,
-0.0678190067410469,
0.070227712392807,
-0.13085584342479706,
-0.037099067121744156,
0.003860368160530925,
0.016657402738928795,
0.05295974016189575,
0.003747355192899704,
-0.0025858685839921236,
0.07462678104639053,
0.034742... | -0.021909 |
shape). HERMES\_DASHBOARD\_PORTAL\_URL is # optional — defaults to production. HERMES\_DASHBOARD\_OAUTH\_CLIENT\_ID=agent:test \ hermes dashboard --host 0.0.0.0 # Or the equivalent via config.yaml (recommended for local dev / on-prem): # # dashboard: # oauth: # client\_id: agent:test # # then just: hermes dashboard --h... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.053029149770736694,
0.06919382512569427,
-0.05054713413119316,
-0.04113934934139252,
0.004525699652731419,
-0.07972823828458786,
-0.05641314014792442,
-0.011341835372149944,
0.053991615772247314,
0.004275825340300798,
-0.023040154948830605,
-0.02083483338356018,
0.06563502550125122,
-0.... | 0.043908 |
Never expose it to the open internet. Put it behind a VPN. [Tailscale](https://tailscale.com/) is the clean option: bind to the machine's tailscale IP (`--host `) and use `http://:9119` as the Remote URL. Only devices on your tailnet can reach it. ::: ### In Hermes Desktop \*\*Settings → Gateway → Remote gateway:\*\* -... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.017350934445858,
0.04053429886698723,
0.04030721262097359,
-0.06347386538982391,
-0.048840269446372986,
-0.0483023077249527,
-0.0745893120765686,
0.06879639625549316,
0.025116680189967155,
0.0000644268438918516,
-0.01798849366605282,
-0.00009286974091082811,
0.032780859619379044,
-0.048... | -0.017501 |
the update skips the frontend build and `hermes dashboard` will build it on first launch. ## Themes & plugins The dashboard ships with six built-in themes and can be extended with user-defined themes, plugin tabs, and backend API routes — all drop-in, no repo clone needed. \*\*Switch themes live\*\* from the header bar... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/web-dashboard.md | main | hermes-agent | [
-0.0499759279191494,
-0.03478337824344635,
0.049337372183799744,
-0.04659479483962059,
0.054625533521175385,
0.02716078981757164,
-0.023833634331822395,
-0.08632792532444,
0.027782831341028214,
0.05594431608915329,
-0.03230005130171776,
0.034703318029642105,
-0.012054335325956345,
-0.04580... | 0.029743 |
# API Server The API server exposes hermes-agent as an OpenAI-compatible HTTP endpoint. Any frontend that speaks the OpenAI format — Open WebUI, LobeChat, LibreChat, NextChat, ChatBox, and hundreds more — can connect to hermes-agent and use it as a backend. Your agent handles requests with its full toolset (terminal, f... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/api-server.md | main | hermes-agent | [
-0.030723411589860916,
-0.03033432736992836,
-0.06875533610582352,
-0.030195320025086403,
-0.011361343786120415,
-0.10676444321870804,
-0.08523101359605789,
-0.018001506105065346,
0.060653988271951675,
0.007700839079916477,
-0.02645106427371502,
-0.006832878105342388,
0.019916806370019913,
... | 0.163315 |
"response", "status": "completed", "model": "hermes-agent", "output": [ {"type": "function\_call", "name": "terminal", "arguments": "{\"command\": \"ls\"}", "call\_id": "call\_1"}, {"type": "function\_call\_output", "call\_id": "call\_1", "output": "README.md src/ tests/"}, {"type": "message", "role": "assistant", "con... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/api-server.md | main | hermes-agent | [
-0.030997173860669136,
0.07356691360473633,
-0.04106629639863968,
0.051153477281332016,
0.007938901893794537,
-0.04431683570146561,
0.010726867243647575,
0.04465959966182709,
0.01980293169617653,
0.0016720666317269206,
-0.02635030634701252,
-0.03178982064127922,
0.09686130285263062,
0.0417... | 0.058362 |
for polling and UI reconciliation. ### GET /v1/runs/\{run\_id\}/events Server-Sent Events stream of the run's tool-call progress, token deltas, and lifecycle events. Designed for dashboards and thick clients that want to attach/detach without losing state. ### POST /v1/runs/\{run\_id\}/stop Interrupt a running agent tu... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/api-server.md | main | hermes-agent | [
-0.11999781429767609,
0.037818148732185364,
-0.0399409718811512,
0.012221201322972775,
-0.013913464732468128,
-0.07953174412250519,
-0.028488216921687126,
-0.039046160876750946,
0.03244846686720848,
0.01940496638417244,
-0.030154502019286156,
0.015355418436229229,
0.030334575101733208,
-0.... | 0.187505 |
WebUI need a stable per-channel identifier for long-term memory (Honcho, etc.) that is \*\*independent\*\* of the transcript-scoped `X-Hermes-Session-Id` (which rotates on `/new`). Pass `X-Hermes-Session-Key` on `/v1/chat/completions`, `/v1/responses`, or `/v1/runs` and Hermes threads it through to `AIAgent(gateway\_se... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/api-server.md | main | hermes-agent | [
-0.07296628504991531,
0.002911351388320327,
-0.054723404347896576,
-0.000017512331396574154,
-0.06424523890018463,
-0.07552161067724228,
0.0680229589343071,
-0.023920804262161255,
0.03288131207227707,
-0.0005781451473012567,
0.05041344836354256,
-0.019765648990869522,
0.025846950709819794,
... | 0.152476 |
ChatBox | 39k | API Host setting | | Jan | 26k | Remote model config | | HF Chat-UI | 8k | OPENAI\_BASE\_URL | | big-AGI | 7k | Custom endpoint | | OpenAI Python SDK | — | `OpenAI(base\_url="http://localhost:8642/v1")` | | curl | — | Direct HTTP requests | ## Multi-User Setup with Profiles To give multiple users their ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/api-server.md | main | hermes-agent | [
0.01629914529621601,
-0.011354424059391022,
-0.07089106738567352,
0.013737909495830536,
-0.050602834671735764,
-0.11088100075721741,
-0.007641766220331192,
0.03826608136296272,
-0.03201156482100487,
0.007909603416919708,
-0.0013726481702178717,
-0.08574379235506058,
0.07917656004428864,
-0... | 0.053548 |
# Features Overview Hermes Agent includes a rich set of capabilities that extend far beyond basic chat. From persistent memory and file-aware context to browser automation and voice conversations, these features work together to make Hermes a powerful autonomous assistant. :::tip Don't know where to start? `hermes setu... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/overview.md | main | hermes-agent | [
-0.07002843916416168,
-0.014938753098249435,
-0.04232202097773552,
-0.021157873794436455,
-0.014817275106906891,
-0.06946635246276855,
-0.008781058713793755,
-0.018091624602675438,
0.005017924588173628,
0.021970301866531372,
-0.0023816830944269896,
0.032993383705616,
0.04032183438539505,
0... | 0.227901 |
(free), ElevenLabs, OpenAI TTS, MiniMax, Mistral Voxtral, Google Gemini, xAI, NeuTTS, KittenTTS, and Piper — plus custom command providers for any local TTS CLI. ## Integrations - \*\*[MCP Integration](mcp.md)\*\* — Connect to any MCP server via stdio or HTTP transport. Access external tools from GitHub, databases, fil... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/overview.md | main | hermes-agent | [
-0.07167500257492065,
-0.06245388835668564,
0.00951736606657505,
-0.05873740464448929,
-0.01348900981247425,
-0.06695298850536346,
-0.06058892607688904,
0.03624578192830086,
-0.026838252320885658,
0.03522660210728645,
-0.011282486841082573,
-0.033998697996139526,
0.06249730661511421,
-0.08... | 0.227382 |
# ACP Editor Integration Hermes Agent can run as an ACP server, letting ACP-compatible editors talk to Hermes over stdio and render: - chat messages - tool activity - file diffs - terminal commands - approval prompts - streamed thinking / response chunks ACP is a good fit when you want Hermes to behave like an editor-n... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/acp.md | main | hermes-agent | [
-0.07298573851585388,
0.005355712026357651,
-0.0517452135682106,
0.004108359571546316,
-0.023532480001449585,
-0.03552579879760742,
-0.039082493633031845,
0.045615263283252716,
0.03389211371541023,
0.0591297447681427,
0.012745073065161705,
-0.03593597933650017,
0.023356446996331215,
-0.016... | 0.143932 |
agent server in Zed settings: ```json { "agent\_servers": { "hermes-agent": { "type": "custom", "command": "hermes", "args": ["acp"] } } } ``` ### JetBrains Use an ACP-compatible plugin and point it at: ```text /path/to/hermes-agent/acp\_registry ``` ## Registry manifest The source copy of Hermes' official ACP Registry... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/acp.md | main | hermes-agent | [
-0.1116015762090683,
0.015545841306447983,
-0.08529294282197952,
-0.02547362819314003,
-0.017144767567515373,
-0.03151823207736015,
0.024627655744552612,
0.016339074820280075,
0.006211135536432266,
-0.03475639596581459,
0.10940175503492355,
0.01472154539078474,
0.006853148341178894,
0.0313... | 0.053919 |
semantics — `allow\_always` writes a permanent allowlist entry the same way the CLI does, while `allow\_session` only affects the in-process approval cache for the current ACP session. ## Troubleshooting ### ACP agent does not appear in the editor Check: - In Zed, open the ACP Registry with `zed: acp registry` and sear... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/acp.md | main | hermes-agent | [
-0.08163566887378693,
0.058974068611860275,
-0.06473053991794586,
-0.04935029149055481,
-0.004244025330990553,
-0.03772807866334915,
0.01693672314286232,
-0.0303261186927557,
-0.018580377101898193,
0.026864903047680855,
0.07208243012428284,
0.0013505322858691216,
0.007604506798088551,
0.04... | 0.031364 |
# Code Execution (Programmatic Tool Calling) The `execute\_code` tool lets the agent write Python scripts that call Hermes tools programmatically, collapsing multi-step workflows into a single LLM turn. The script runs in a child process on the agent host, communicating with Hermes over a Unix domain socket RPC. ## How... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/code-execution.md | main | hermes-agent | [
-0.06958463788032532,
-0.015989145264029503,
-0.08290070295333862,
0.0148020563647151,
-0.024407656863331795,
-0.14137141406536102,
-0.0686422735452652,
0.002763396129012108,
0.037627313286066055,
0.0054033915512263775,
0.00229816185310483,
0.03727243095636368,
0.06920237839221954,
-0.0727... | 0.097641 |
`open(".env")` to work the same way they do in `terminal()`. This is almost always what you want. \*\*When to flip to `strict`:\*\* you need maximum reproducibility — you want the same interpreter every session regardless of which venv the user activated, and you want scripts quarantined from the project tree (no risk ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/code-execution.md | main | hermes-agent | [
-0.02898053638637066,
-0.0012201275676488876,
-0.017241543158888817,
-0.057976823300123215,
0.04152265563607216,
-0.10211193561553955,
-0.029123447835445404,
0.05749364197254181,
0.03409606218338013,
0.03757866099476814,
-0.027940282598137856,
0.0023657132405787706,
0.03382076323032379,
0.... | -0.014977 |
skill declares `required\_environment\_variables` in its frontmatter, those variables are \*\*automatically passed through\*\* to both `execute\_code` and `terminal` child processes after the skill is loaded. This lets skills use their declared API keys without weakening the security posture for arbitrary code. For non... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/code-execution.md | main | hermes-agent | [
-0.003690870711579919,
-0.00289233960211277,
-0.0930878147482872,
-0.0006236046901904047,
-0.05107811093330383,
-0.042143288999795914,
0.06197873875498772,
0.023146122694015503,
-0.0652836486697197,
0.0307177547365427,
0.026911849156022072,
-0.02168162725865841,
0.10410242527723312,
0.0216... | 0.053924 |
Windows — the agent falls back to regular sequential tool calls. | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/code-execution.md | main | hermes-agent | [
-0.06714871525764465,
-0.005976735148578882,
-0.04424884915351868,
0.027379289269447327,
-0.06864809989929199,
0.016456453129649162,
0.007450990844517946,
-0.014567892998456955,
0.059420395642519,
-0.0220778975635767,
0.01783027872443199,
0.09692812711000443,
-0.04170848801732063,
-0.03244... | 0.154916 |
# Language Server Protocol (LSP) Hermes runs full language servers — pyright, gopls, rust-analyzer, typescript-language-server, clangd, and ~20 more — as background subprocesses and feeds their semantic diagnostics into the post-write lint check used by `write\_file` and `patch`. When the agent edits a file, it sees ex... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/lsp.md | main | hermes-agent | [
-0.059031639248132706,
-0.08508472889661789,
0.027771232649683952,
0.017267514020204544,
0.05563371255993843,
-0.05685831233859062,
-0.009004269726574421,
0.03273693472146988,
0.074471615254879,
0.037652306258678436,
0.03340933099389076,
0.04463707655668259,
-0.00029828105471096933,
0.0054... | 0.192329 |
For "manual" entries, install the server through whatever toolchain manager makes sense for that language (rustup, ghcup, opam, brew, …). Hermes auto-detects the binary on PATH or in `/lsp/bin/`. A few servers are installed alongside a peer dependency that npm won't auto-pull. The current case is `typescript-language-s... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/lsp.md | main | hermes-agent | [
-0.04671500623226166,
-0.08462191373109818,
0.028240077197551727,
-0.036142461001873016,
0.03253922611474991,
-0.03374653682112694,
-0.009487657807767391,
0.04527562856674194,
-0.052410729229450226,
-0.003431913675740361,
-0.024072227999567986,
-0.021697474643588066,
-0.01755448803305626,
... | 0.118237 |
syntax check (`ast.parse` for Python, `json.loads` for JSON, etc.) which ships unchanged from earlier versions. To disable a single language without disabling the whole layer: ```yaml lsp: servers: rust-analyzer: disabled: true ``` ## Troubleshooting \*\*`hermes lsp status` shows a server as "missing"\*\* The binary is... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/lsp.md | main | hermes-agent | [
-0.022227836772799492,
-0.002703276928514242,
-0.0017012415919452906,
0.01816048100590706,
0.07722224295139313,
-0.0845547616481781,
-0.00604067649692297,
-0.004818955902010202,
-0.02908681146800518,
-0.061266086995601654,
0.06784293055534363,
-0.022112425416707993,
0.010955242440104485,
0... | 0.086521 |
# Batch Processing Batch processing lets you run the Hermes agent across hundreds or thousands of prompts in parallel, generating structured trajectory data. This is primarily used for \*\*training data generation\*\* — producing ShareGPT-format trajectories with tool usage statistics that can be used for fine-tuning o... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/batch-processing.md | main | hermes-agent | [
-0.000922309176530689,
-0.012999762780964375,
-0.08412114530801773,
0.004527341108769178,
-0.024092359468340874,
-0.06871771812438965,
-0.04989096522331238,
-0.0007767239003442228,
-0.04073944315314293,
0.023815035820007324,
-0.006638364866375923,
-0.018792415037751198,
0.025999916717410088,... | 0.111906 |
in log previews (default: 100) | | `--prefill\_messages\_file` | Path to JSON file with prefill messages for few-shot priming | ## Toolset Distributions Each prompt gets a randomly sampled set of toolsets from a \*\*distribution\*\*. This ensures training data covers diverse tool combinations. Use `--list\_distribution... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/batch-processing.md | main | hermes-agent | [
-0.039267584681510925,
-0.05326457694172859,
-0.03477935865521431,
0.027233697474002838,
0.03332074359059334,
-0.056966472417116165,
-0.004651205148547888,
0.008603762835264206,
0.04213466867804527,
-0.00391143886372447,
0.024711020290851593,
-0.01669665053486824,
0.012091252021491528,
-0.... | 0.04254 |
--dataset\_file=data/eval\_suite.jsonl \ --batch\_size=10 \ --run\_name=eval\_gpt4 \ --model=openai/gpt-4o \ --num\_workers=4 \ --max\_turns=10 ``` ### Per-Prompt Container Images For benchmarks requiring specific environments, each prompt can specify its own container image: ```jsonl {"prompt": "Install numpy and comp... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/batch-processing.md | main | hermes-agent | [
0.027818946167826653,
0.061959441751241684,
-0.06064383685588837,
0.02832302451133728,
0.08537892252206802,
-0.08191878348588943,
-0.02970796264708042,
0.07982508093118668,
-0.0440235398709774,
-0.01949964091181755,
-0.02791306935250759,
-0.0932316854596138,
0.05983110889792442,
0.05505214... | 0.06713 |
# Browser Automation Hermes Agent includes a full browser automation toolset with multiple backend options: - \*\*Browserbase cloud mode\*\* via [Browserbase](https://browserbase.com) for managed cloud browsers and anti-bot tooling - \*\*Browser Use cloud mode\*\* via [Browser Use](https://browser-use.com) as an altern... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
-0.13004997372627258,
0.018061188980937004,
-0.03255108371376991,
-0.008272613398730755,
0.08739253878593445,
-0.0880437046289444,
-0.03302207589149475,
-0.07261036336421967,
-0.028716521337628365,
-0.05192152038216591,
0.05049924552440643,
-0.04125094786286354,
0.054241444915533066,
-0.03... | 0.009856 |
``` With auto-routing disabled, private URLs are rejected with `"Blocked: URL targets a private or internal address"` unless you also set `browser.allow\_private\_urls: true` (which lets the cloud provider attempt them — usually won't work since Browserbase etc. can't reach your LAN). Requirements: the local sidecar us... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
0.007708520628511906,
0.006099986843764782,
-0.011452610604465008,
-0.018665337935090065,
0.04805727303028107,
-0.06952559947967529,
-0.035992927849292755,
-0.02494226209819317,
-0.06087665632367134,
-0.016991419717669487,
0.07489389181137085,
0.020446591079235077,
-0.0010255200322717428,
... | -0.034881 |
up. :::warning Nested path matters Hermes reads `browser.camofox.managed\_persistence`, \*\*not\*\* a top-level `managed\_persistence`. A common mistake is writing: ```yaml # ❌ Wrong — Hermes ignores this managed\_persistence: true ``` If the flag is placed at the wrong path, Hermes silently falls back to a random ephe... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
-0.03412501886487007,
0.03463619202375412,
-0.07865423709154129,
0.015749048441648483,
0.0524885468184948,
-0.09596226364374161,
0.045940496027469635,
-0.029749715700745583,
0.01714588701725006,
-0.055123403668403625,
0.006706646643579006,
0.02231564372777939,
0.09622786939144135,
0.023507... | 0.046542 |
issues `GET /tabs?userId=` (5-second timeout). 2. If any tab in the response has `listItemId == session\_key`, Hermes adopts the most recently created one in that group. 3. Otherwise, Hermes adopts the most recently created tab for the user (any `listItemId`). 4. If no tabs exist or the request fails, Hermes falls back... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
-0.05948225036263466,
-0.008093049749732018,
-0.01526628341525793,
-0.005059915129095316,
0.015475361607968807,
-0.11502011120319366,
0.06085750088095665,
-0.0008583684684708714,
0.04577161744236946,
0.03763994947075844,
0.0023725253995507956,
-0.020532645285129547,
0.027810687199234962,
0... | 0.067389 |
--user-data-dir="$HOME/.hermes/chrome-debug" \ --no-first-run \ --no-default-browser-check & ``` Then launch the Hermes CLI and run `/browser connect`. \*\*Why `--user-data-dir`?\*\* Without it, launching a Chromium-family browser while a regular instance is already running typically opens a new window on the existing ... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
0.007295738905668259,
-0.0021218289621174335,
-0.04070054367184639,
-0.020070930942893028,
0.02531513012945652,
-0.11508836597204208,
-0.07577092945575714,
0.03998976945877075,
-0.04844290018081665,
-0.05298512428998947,
0.08798012137413025,
-0.0179936271160841,
-0.004058520309627056,
0.02... | -0.042685 |
element identified by its ref ID from the snapshot. ``` Click @e5 to press the "Sign In" button ``` ### `browser\_type` Type text into an input field. Clears the field first, then types the new text. ``` Type "hermes agent" into the search field @e3 ``` ### `browser\_scroll` Scroll the page up or down to reveal more co... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
-0.06647244095802307,
0.02216600999236107,
-0.024823445826768875,
0.03677031397819519,
0.09632554650306702,
-0.03656037524342537,
-0.0069767385721206665,
-0.03410109505057335,
0.0049686189740896225,
-0.00818265788257122,
0.029579762369394302,
0.042654551565647125,
0.1327131688594818,
-0.05... | 0.138676 |
is independent — sessions do not persist between calls. \*\*Cross-origin iframes:\*\* pass `frame\_id` (from `browser\_snapshot.frame\_tree.children[]` where `is\_oopif=true`) to route the CDP call through the supervisor's live session for that iframe. This is how `Runtime.evaluate` inside a cross-origin iframe works o... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
-0.01803627237677574,
0.008032291196286678,
-0.05278285592794418,
0.006583591923117638,
0.0701029822230339,
-0.09088578820228577,
-0.01232969667762518,
0.041019998490810394,
0.008457519114017487,
-0.04708962142467499,
-0.03214408457279205,
-0.01034554559737444,
-0.08977194130420685,
0.0545... | 0.045091 |
# default: false ``` When enabled, recording starts automatically on the first `browser\_navigate` and saves to `~/.hermes/browser\_recordings/` when the session closes. Works in both local and cloud (Browserbase) modes. Recordings older than 72 hours are automatically cleaned up. ## Stealth Features Browserbase provid... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/browser.md | main | hermes-agent | [
-0.08037126064300537,
-0.016709621995687485,
-0.0027212328277528286,
0.0017040111124515533,
0.07808203995227814,
-0.04537194222211838,
-0.03335525467991829,
-0.059038665145635605,
-0.06191143020987511,
-0.01601547934114933,
0.061950407922267914,
0.021924100816249847,
0.016342075541615486,
... | 0.080029 |
# MCP (Model Context Protocol) MCP lets Hermes Agent connect to external tool servers so the agent can use tools that live outside Hermes itself — GitHub, databases, file systems, browser stacks, internal APIs, and more. If you have ever wanted Hermes to use a tool that already exists somewhere else, MCP is usually the... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/mcp.md | main | hermes-agent | [
-0.041272878646850586,
-0.020495660603046417,
-0.024503404274582863,
-0.03436913713812828,
-0.008747301064431667,
-0.07293344289064407,
-0.04317726567387581,
0.010443279519677162,
-0.018266241997480392,
0.03189223259687424,
0.01634933613240719,
-0.03338523954153061,
0.05992048978805542,
-0... | 0.10458 |
select everything, no filter is written (cleanest config shape, identical behavior). \*\*If the probe fails\*\* (server unreachable, OAuth not yet completed, backing service not running), the install still succeeds: the manifest's `tools.default\_enabled` is applied directly (if declared), or no filter is written (if n... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/mcp.md | main | hermes-agent | [
-0.004109840840101242,
-0.0034268954768776894,
0.0446932315826416,
-0.014760276302695274,
0.050456516444683075,
-0.11586333066225052,
-0.027496764436364174,
-0.05605430155992508,
-0.009830198250710964,
0.014965224079787731,
0.055260203778743744,
-0.007337873335927725,
0.03623528033494949,
... | 0.048107 |
OAuth callback on a local loopback port. Tokens are cached at `~/.hermes/mcp-tokens/.json` with 0o600 perms; subsequent runs reuse them silently until refresh fails. \*\*Remote / headless hosts.\*\* When Hermes runs on a different machine than your browser, the loopback callback can't reach your laptop. Two ways to com... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/mcp.md | main | hermes-agent | [
-0.00036286842077970505,
0.02867361716926098,
-0.006505889818072319,
-0.010061538778245449,
0.010443033650517464,
-0.10505897551774979,
-0.11134838312864304,
0.010865242220461369,
0.0244306568056345,
-0.017862088978290558,
0.014002983458340168,
0.04096461459994316,
0.06343794614076614,
-0.... | -0.000123 |
look up the command and args. The preset only supplies defaults — anything else (env vars, headers, filtering) you pass on the same command line still wins. | Preset | What it wires up | |---|---| | `codex` | The Codex CLI's MCP server (`codex mcp-server` over stdio). Requires the `codex` CLI on PATH. | ```bash # Add C... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/mcp.md | main | hermes-agent | [
-0.027203528210520744,
-0.0402904637157917,
-0.006402450148016214,
-0.05031070485711098,
-0.07485345751047134,
-0.055329736322164536,
0.005690207704901695,
0.03287840262055397,
-0.013268360868096352,
0.012607106938958168,
0.01777106150984764,
-0.03775450214743614,
0.02066863887012005,
-0.0... | 0.100138 |
sending a `notifications/tools/list\_changed` notification. When Hermes receives this notification, it automatically re-fetches the server's tool list and updates the registry — no manual `/reload-mcp` required. This is useful for MCP servers whose capabilities change dynamically (e.g. a server that adds tools when a n... | https://github.com/NousResearch/hermes-agent/blob/main/website/docs/user-guide/features/mcp.md | main | hermes-agent | [
-0.03952831029891968,
-0.08074131608009338,
0.06488426774740219,
0.029932120814919472,
0.04320930689573288,
-0.08139719069004059,
0.015553969889879227,
-0.07840664684772491,
0.05682172253727913,
0.037796374410390854,
0.014694243669509888,
0.011328431777656078,
0.051694076508283615,
-0.0124... | 0.076504 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.