Upload 7 files

.gitignore

@@ -18,3 +18,12 @@ __pycache__/
 # OS generated files
 .DS_Store
 Thumbs.db
+
+# Hackathon research files (not part of submission)
+*.pdf
+hackathon_*.txt
+hackathon_*.pdf
+problem_statement.txt
+
+# Leftover subdirectory
+openenv-git-conflict-resolver/

openenv.yaml

@@ -1,64 +1,6 @@
+spec_version: 1
 name: git-conflict-resolver
-version: "1.0.0"
-description: >
-  An RL environment where AI agents learn to resolve Git merge conflicts.
-  The agent receives a file containing conflict markers (<<<<<<<, =======, >>>>>>>)
-  and must produce a clean, correctly resolved version of the file.
-  Tasks range from resolving a single obvious conflict to combining additive
-  changes from both branches in a semantically correct way.
-
-author: "Agent Smith"
-license: MIT
-tags:
-  - openenv
-  - git
-  - code-review
-  - software-engineering
-  - conflict-resolution
-
-tasks:
-  - name: single_conflict
-    difficulty: easy
-    description: "Resolve a single Git conflict block by accepting the correct incoming change."
-  - name: multi_conflict
-    difficulty: medium
-    description: "Resolve three conflict blocks, each requiring a different resolution strategy."
-  - name: logic_conflict
-    difficulty: hard
-    description: "Resolve two conflict blocks by combining additive changes from both branches."
-
-action_space:
-  type: text
-  description: "The agent outputs the full resolved file content as a string."
-
-observation_space:
-  type: structured
-  fields:
-    - task_name: string
-    - task_description: string
-    - filename: string
-    - file_language: string
-    - conflicted_content: string
-    - branch_ours: string
-    - branch_theirs: string
-    - num_conflicts: integer
-    - last_attempt: string (nullable)
-    - last_error: string (nullable)
-    - step: integer
-    - max_steps: integer
-    - done: boolean
-
-reward:
-  type: float
-  range: [0.0, 1.0]
-  shaped: true
-  description: >
-    Shaped reward per step. Components include:
-    improvement bonus (score delta), marker-free bonus,
-    perfect-match bonus, stagnation penalty, and step cost.
-
-endpoints:
-  reset: POST /reset
-  step: POST /step
-  state: GET /state
-  health: GET /health
+type: space
+runtime: fastapi
+app: server.app:app
+port: 7860