1ba56fd7e3
Add Skill: three-tier-evaluation-pipeline
...
Extracted from: https://github.com/itszhaoziyan-n/AgentKit.git
Score: 1.0
2026-08-05 17:05:03 +00:00
d81ddeda88
Scout window: 6 months (Feb-Aug 2026)
...
- pushed_after: 2026-02-01 (was 2024-06-01)
- Keeps scope tight on recent, relevant repos
2026-08-05 16:44:22 +00:00
a14f09bec2
Add 2 new skills + dedup fix
...
New skills:
- langgraph-workflow-creation (from LangGraphProjects)
- agent-supervisor (from multi_agent_workflow_demo_in_langgraph)
Fixes:
- Publisher dedup: skip existing skills
- Older repos: pushed_after 2024-06-01 (was 2026-05-01)
- Lower stars: 10 (was 15)
2026-08-05 15:55:36 +00:00
7f496feb90
Publisher dedup check + run.py skip display
...
- Skip skills already in skills/ directory (no duplicate PRs)
- Run.py shows SKIP status with reason
- Fixed: was re-publishing same 5 skills every run
2026-08-05 15:49:57 +00:00
5f917f4121
Add Publisher v2 + 5 extracted skills
...
Publisher fixes:
- Checkout new branch before push (was pushing main ref)
- Verify files staged before commit
- Handle duplicate files gracefully
- Clean error reporting per stage
Skills merged to main:
- mcp-server-setup (from pipeshub-ai)
- research-pipeline (from Blacknode)
- multi-agent-sequential-workflow (from Fast-LLM-Agent-MCP)
- unifai-workflow-execution (from UnifAI)
- code-review-agent-workflow (from AgentKit)
2026-08-05 15:15:09 +00:00
09b62adb93
Remove __pycache__ and runs/ from tracking
2026-08-05 15:00:08 +00:00
dc40d4c0db
Pipeline v2: deterministic reviewer, implementation extraction, publisher fix
...
- Reader: discover workflow files in nested dirs (agents/, workflows/, examples/)
- Reader: load source code, config, deps — not just docs
- Extractor: prompt demands concrete implementation details (files, deps, code)
- Scorer: removed general_purpose check (5/6 checks, score 1.0)
- Generator: includes Setup, Key Files, Implementation Details sections
- Reviewer: replaced LLM review with 8 deterministic structural checks
- Publisher: handle 409 duplicate PR gracefully as success
- 5 skills published as PRs #6-#10 on Gitea
2026-08-05 13:58:38 +00:00
8da8d703da
Agent Skills Pipeline v1.0
...
8-stage pipeline: Scout → Filter → Reader → Extractor → Score → Generator → Reviewer → Publisher
- Scout: GitHub search with token auth + rate limit retry
- Filter: Deterministic rules (language, stars, age, keywords)
- Reader: Incremental context loading (README → docs → examples → code)
- Extractor: LLM workflow extraction with JSON retry
- Score: Rule-based evaluation (no LLM)
- Generator: Standardized Hermes Skill format
- Reviewer: Independent LLM review (separate from generator)
- Publisher: Branch + PR to Gitea
First run: 5 repos discovered, 0 extracted (correct — all frameworks, no workflows)
2026-08-05 05:51:06 +00:00
Hermes Pipeline