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.
npx agentmods add instructions/srnichols/plan-forge/self-repair-reportinggit clone --depth 1 https://github.com/srnichols/plan-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/instructions/srnichols/plan-forge/self-repair-reporting)<a href="https://agentmods.dev/instructions/srnichols/plan-forge/self-repair-reporting"><img src="https://agentmods.dev/badge/instructions/srnichols/plan-forge/self-repair-reporting.svg" alt="Measured on agentmods" 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 | $0.01327 | $0.01327 |
| Opus 5 | $0.00664 | $0.00664 |
| Sonnet 5 | $0.00265 | $0.00265 |
| Haiku 4.5 | $0.00133 | $0.00133 |
Grade A, and why
plan-forge self-repair-reporting.instructions.md 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 4d 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 — 127 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Self-Repair Reporting
When to use: You discovered a defect in Plan Forge itself (plan files, orchestrator, CLI, prompts, instruction files) during execution and worked around it. File a meta-bug so the fix is captured.
Two Lanes at a Glance
| Project Bug | Meta Bug (Self-Repair) | |
|---|---|---|
| What | Bug in the code you're building | Bug in Plan Forge itself |
| Tool | forge_bug_file / tempering |
forge_meta_bug_file |
| Target repo | Workspace git remote |
.forge.json#meta.selfRepairRepo → fallback srnichols/plan-forge |
| Labels | bug, severity |
self-repair, class label |
If the defect is in user project code, use forge_bug_file. If the defect is in Plan Forge's own plans, orchestrator, CLI, or prompts — use forge_meta_bug_file.
When to Fire forge_meta_bug_file
Fire the tool when you had to work around one of these three defect classes:
1. Plan Defect (plan-defect)
The plan file was wrong. Brittle validation gate, missing scope entry, wrong file path, over-narrow grep pattern, unsatisfiable dependency. You edited the plan or worked around the gate to keep moving.
2. Orchestrator / CLI Defect (orchestrator-defect)
Runtime bug in Plan Forge. Gate timeout too short, spawn failure on Windows, stash not popped after failed retry, estimator recommends a CLI-less model. You worked around it or the retry loop papered over it.
3. Prompt / Template Defect (prompt-defect)
A step-N prompt emitted unsafe output, an instruction file is missing a rule that would have prevented the problem, or a template placeholder wasn't expanded.
Worked Examples
Example 1: Brittle grep gate — plan-defect
Phase-28.2 Slice 2 — The plan gate used grep -q 'A' file || grep -q 'B' file but the worker chose identifier isApiOnlyModel, which matched neither pattern. The agent edited the plan gate inline to broaden the match.
{
"class": "plan-defect",
"title": "Phase-28.2 Slice 2 gate too narrow for chosen identifier",
"symptom": "grep gate did not match worker's chosen export name",
"workaround": "Edited plan gate to accept isApiOnlyModel",
"slice": "2",
"plan": "docs/plans/Phase-28.2-PLAN.md"
}
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.
- 4d ago First seen · 127 lines · 1,327 tokens per session scan A 76cbc57d3934
plan-forge self-repair-reporting.instructions.md is an instructions file published in the GitHub repository srnichols/plan-forge (5 stars, last pushed yesterday), licensed MIT. It adds 1,327 tokens to every session, about $0.0066 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 2026-08-31.
Other instructions, from other repositories
skillfoundry copilot-instructions.md
Copilot instructions for samibs/skillfoundry, covering github copilot — skillfoundry project instructions, identity, non-negotiable rules, code quality and security.
skillfoundry AGENTS.md
AGENTS.md instructions for samibs/skillfoundry, covering what this is, philosophy, how to use skills, available skills by category and core workflow.
arai CLAUDE.md
Instructions for taniwhaai/arai, covering claude.md — arai, commands, architecture, prompt-collector module (src/promptcollector.rs) and extending the match pipeline.
ai-control-framework CLAUDE.md
Instructions for sgharlow/ai-control-framework, covering claude code configuration, framework status (v2.0) — verified 2026-07-18, ⚠️ critical: implementation required, mandatory session start procedure and hard stop conditions.
arai AGENTS.md
Instructions for taniwhaai/arai, covering agents.md — arai, core discipline (non-negotiable), taniwha / subagent rules, work style and tool usage.
skillfoundry CLAUDE.md
Claude Code instructions for samibs/skillfoundry, covering claude.md — skillfoundry framework project instructions, philosophy, mandatory production rules, ui/layout rules and documentation rules.