ownex-forge
Skill1 starsChanged 3 months agoMITView on GitHub
Goal: Process incoming bounty reports — triage submissions, validate scope, verify evidence, and prepare reward recommendations. App: forge · Cadence: 2h · Risk: medium discover → triage → classify → verify → act → notify
In this repository8 other files it gives its agents.
- AGENTS.mdAGENTS.md
- graftSkill
- loop-triageSkill
- ownex-atlasSkill
- ownex-odysseySkill
- ownex-pulseSkill
- ownex-securitySkill
- ownex-vaultSkill
The file
Files
OWNEX Forge Cycle — Bounty Forge
Goal: Process incoming bounty reports — triage submissions, validate scope, verify evidence, and prepare reward recommendations.
App: forge · Cadence: 2h · Risk: medium
Default Phases
discover → triage → classify → verify → act → notify
Human Gates
- High-reward bounties (>$1000)
- Out-of-scope determinations
- Dispute resolution
Required Skills
bounty-triage— Triage incoming bounty submissionsscope-check— Validate target scope and program rulesevidence-verify— Verify reproducibility and evidence qualityreward-calc— Calculate reward based on severity + program
State File
# Forge Cycle State
Last run: <timestamp>
Pending Review:
- B-042: Auth bypass on api.example.com — reward $500
Watch List:
- B-038: Info disclosure — awaiting user response (48h)
Escalated:
- B-040: SSRF with PII exposure — legal reviewSuccess Metrics
- Time from submission to triage completion
- % of submissions accepted / rejected
- Average reward accuracy (vs final payout)
- Submitter satisfaction score
Budget
- Daily cap: 75k tokens
- Max runs: 6/day (every 2h during active hours)
- L1: report only
- L2: auto-reward calculation (with human approval gate)
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.