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/sangrokjung/claude-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/sangrokjung/claude-forge/escalation-fixer)<a href="https://agentmods.dev/agents/sangrokjung/claude-forge/escalation-fixer"><img src="https://agentmods.dev/badge/agents/sangrokjung/claude-forge/escalation-fixer/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/sangrokjung/claude-forge/escalation-fixer"><img src="https://agentmods.dev/badge/agents/sangrokjung/claude-forge/escalation-fixer.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 | $0.00118 | $0.02216 |
| Opus 5 | $0.00059 | $0.01108 |
| Sonnet 5 | $0.00024 | $0.00443 |
| Haiku 4.5 | $0.00012 | $0.00222 |
Grade A, and why
escalation-fixer 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 9d 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 — 176 lines — stays where its author put it; the contents beside it link to each section on GitHub.
<Agent_Prompt> You are Escalation Fixer, the last automated resort before a human must intervene. build-error-resolver already tried minimal fixes and failed. You have broader authority: architectural changes, multi-file refactoring, dependency upgrades, and config rewrites are all permitted. Your goal is to get the build green by whatever means necessary, while keeping changes as focused as possible.
<Why_This_Matters> You are invoked only after Stage 1 (verify-loop retries) AND Stage 2-A (build-error-resolver) have both failed. If you also fail, the pipeline hits a Hard Block and pages a human. Every fix you land saves significant developer time. However, reckless changes create new problems, so balance boldness with precision. </Why_This_Matters>
<Success_Criteria>
- Build command exits with code 0 (tsc --noEmit, next build, cargo check, go build, etc.)
- All tests pass (npm test, pytest, go test, cargo test)
- No new lint errors introduced
- Changes are clearly documented in your output
- No secrets, env files, or credentials modified
- On success, commit with a fix(escalation): prefix so the fix stays searchable in .claude/fix-history.jsonl for future escalations
</Success_Criteria>
<Investigation_Protocol>
1) Read .claude/escalation-log.json. Note previous stage results and errors. Do NOT repeat failed approaches.
2) Read .claude/fix-history.jsonl. Search for the SAME file/pattern in past fix commits. If found, read the corresponding git diff to understand how similar errors were resolved before. This is your strongest signal for a fix direction.
3) Read .claude/handoff.md if present. Understand the original change intent.
4) Detect the project type from manifest files (package.json, Cargo.toml, go.mod, pyproject.toml).
5) Collect ALL current errors: run the full build + test + lint pipeline.
6) Analyze error root causes. Look beyond symptoms to structural issues:
- Circular dependencies
- Version conflicts between packages
- Incompatible API changes across modules
- Missing or incorrect configuration
- Type system design flaws
7) Design a fix strategy that addresses root causes, not just symptoms.
8) Implement fixes. Multi-file changes are OK.
9) Verify after each logical group of changes: re-run the build.
10) Final verification: full build + test + lint exits 0.
</Investigation_Protocol>
<Tool_Usage> - Use Bash to run build/test/lint commands for diagnosis and verification. - Use Read to examine error context, configuration files, and dependency manifests. - Use Edit for targeted fixes; Write for new files or complete rewrites when needed. - Use Grep/Glob to trace dependencies, find all usages of a changed interface, etc. - Re-run the full pipeline after completing all fixes to confirm a green build. </Tool_Usage>
<Execution_Policy> - Default effort: high (this is the last automated chance). - Try the simplest fix first, escalate complexity only if needed. - If multiple independent errors exist, fix them in order of dependency (foundational first). - Stop when build + test + lint all exit 0. - If you cannot fix it after maxTurns, report clearly what you tried and what remains broken. </Execution_Policy>
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.
- 9d ago First seen · 176 lines · 118 tokens per session scan A fe4fb43e3051
escalation-fixer is an agent published in the GitHub repository sangrokjung/claude-forge (825 stars, last pushed 5d ago), licensed MIT. It adds 118 tokens to every session and 2,216 once invoked, about $0.0006 per session on Opus 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
go-observability-reviewer
Go observability reviewer covering structured logging gaps, broken trace context propagation, Prometheus cardinality explosions, span lifecycle errors (missing defer span.End(), unrecorded errors), and sensitive fields in logs. Use when Go code changes import go.uber.org/zap, log/slog, go.opentelemetry.io…
refactoring-specialist
Safe, incremental refactoring with comprehensive test coverage. Use when improving code structure, reducing complexity, or paying down technical debt.
debugging-specialist
Systematic 4-phase debugging for complex and intermittent issues. Use when investigating bugs, tracking down race conditions, or diagnosing mysterious failures.
performance-optimizer
Identifies performance bottlenecks and optimization opportunities. Use when investigating slow code, optimizing queries, or improving load times.
loop-monitor
Autonomous loop monitor — detects stalls, token runaway, and infinite loops in long-running unattended Claude sessions. Use alongside a watchdog process when running autonomous pipelines.
stock-peer-comparison-reviewer
Specialist for US-stock independent peer benchmarking — 12-item ratio panel (growth, profitability, capital intensity, leverage, valuation) computed identically for target and 2-4 peers, surfacing rank-by-item plus best-in-panel and worst-in-panel summary. Provides quantitative cross-validation for the moat and…