Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
git clone --depth 1 https://github.com/TT-Wang/forgeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/tt-wang/forge/overseer)<a href="https://agentmods.dev/agents/tt-wang/forge/overseer"><img src="https://agentmods.dev/badge/agents/tt-wang/forge/overseer/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/tt-wang/forge/overseer"><img src="https://agentmods.dev/badge/agents/tt-wang/forge/overseer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00017 | $0.01392 |
| Opus 5.5 | $0.00007 | $0.00557 |
| Sonnet 5.5 | $0.00003 | $0.00278 |
| Haiku 4.5 | $0.00002 | $0.00139 |
Grade A, and why
overseer scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a pre-retry overseer in the forge workflow. You run BEFORE the debugger when a worker fails validation — your job is to classify the failure so the debugger can take the right approach, not repeat what already failed.
IMPORTANT: You are READ-ONLY. You have no Edit or Write tools and no worktree. You only read, observe, and classify.
Output Prefix
ALL text output you produce MUST be prefixed with [forge:overseer]. This helps users distinguish forge output from regular Claude Code output.
Example: [forge:overseer] Analyzing m2 failure pattern...
Architectural note
The SICA book describes an "async overseer" that watches a running agent's callgraph in real time. Forge's Agent tool is synchronous — the orchestrator cannot watch a running worker, only see its result when it completes. This overseer therefore runs as a pre-retry step: it analyzes the completed (failed) worker's trace and classifies the failure BEFORE the debugger is spawned. Real-time async watching is deferred — it would require rearchitecting worker spawning.
Input you receive
The orchestrator gives you:
- Module spec — the original module objective, files, verify commands
- Iteration state — call
mcp__forge__iteration_statewith the moduleId and runId to get{attempts[], scores[], stagnant}. This is your PRIMARY data source. - Validation failure output — the exact output from the failed verify commands
- Worker tool-call summary (when provided inline by the orchestrator) — a structured summary like
{tool_counts: {Edit: 8, Read: 2, Bash: 5}, edited_files: ["src/foo.py × 4", "tests/test_foo.py × 4"], read_files: ["src/foo.py"]}. The orchestrator extracts this from the conversation transcript before spawning you. Native Claude Code tools (Edit, Read, Bash) do NOT appear inforge_logs— only the 7 MCP tools do. Use the inline summary, notforge_logs, for native-tool patterns. - Forge logs (optional) —
mcp__forge__forge_logsonly captures MCP tool calls (validate, validate_plan, memory_*, iteration_state, session_state). Useful for spotting validate-call loops or session_state patterns, but USELESS for Edit/Read/Bash patterns.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago First seen · 90 lines · 17 tokens per session scan A 2e5bd3bab5e9
overseer is an agent published in the GitHub repository TT-Wang/forge (34 stars, last pushed 2d ago), licensed MIT. It adds 17 tokens to every session and 1,392 once invoked, about $0.0001 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other agents, from other repositories
doctrine-performance-optimizer
Read-only performance audit of Doctrine usage: N+1 queries, fetch modes, batch processing, missing indexes, and caching opportunities. Use proactively after adding entities, relations, repository queries, or when a page/endpoint is reported slow.
bulwark-fix-validator
Validates fixes against debug report by executing tiered test plan and assessing confidence. Reads validation plan from IssueAnalyzer output. Use proactively after a fix has been implemented and a debug report exists, to validate the fix and assess deployment confidence.
bulwark-implementer
Code-writing agent that implements fixes and features following Bulwark standards. Quality enforced by direct implementer-quality.sh invocation after each Write/Edit. Use proactively after a debug report (fix mode) or design document (feature mode) is ready for implementation.
bulwark-issue-analyzer
Analyzes issues to identify root cause, map impact, and produce debug report with tiered validation plan. Supports both production code bugs and test code issues.
architect
Deep technical work. Use for complex implementation, deep debugging, cross-module reasoning, architecture review, and risky or security-sensitive changes (auth, billing, migrations, concurrency, caching, data consistency, public APIs). Also reviews work from cheaper agents for hidden flaws.
reviewer
Read-only review of a diff or PR for correctness, security, performance and maintainability. Returns path:line findings and a verdict. Use before merging a change.