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/kurone-kito/idd-skill/idd-roadmap-auditgit clone --depth 1 https://github.com/kurone-kito/idd-skillWrote 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/kurone-kito/idd-skill/idd-roadmap-audit)<a href="https://agentmods.dev/instructions/kurone-kito/idd-skill/idd-roadmap-audit"><img src="https://agentmods.dev/badge/instructions/kurone-kito/idd-skill/idd-roadmap-audit.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 5.1 | $0.02405 | $0.02405 |
| Opus 5 | $0.01203 | $0.01203 |
| Sonnet 5 | $0.00481 | $0.00481 |
| Haiku 4.5 | $0.00241 | $0.00241 |
Grade C, and why
idd-skill idd-roadmap-audit.instructions.md scanned grade C with 1 finding 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 6d 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.
Hidden instructionshighPrompt injection
Directives inside HTML comments, invisible characters or bidirectional overrides are read by the model and not by the person reviewing the file.
<!-- idd-generated-from: idd-template/.github/instructions/idd-roadmap-audit.instructions.md Generated by sync-docs. Edit the source above, then run `node scripts/sync-docs.mjs --apply`; do not edit this file. --> How it starts
The opening of the file, as written. The whole thing — 191 lines — stays where its author put it; the contents beside it link to each section on GitHub.
IDD — Roadmap Audit Phase (A1.5)
Read this file after A1 selects an open roadmap and before A2 enumerates child issues.
A1.5 — Audit completed roadmaps
After A1 selects an open roadmap, inspect whether the roadmap appears complete before enumerating more work. This step belongs in Discover because Discover owns roadmap selection and global readiness. F4 remains scoped to the just-merged PR and local cleanup while holding only the child issue claim; it must not close a parent roadmap as a side effect. F5 still loops back here, so the next Discover pass can evaluate parent roadmap state after child PRs merge.
Run the completion audit only when the selected roadmap has explicit child work such as task-list issue references or GitHub sub-issue relationships. If the roadmap has no explicit child work, report that it is childless or malformed and continue to A2; do not close it based on absence of candidates.
Fetch the selected roadmap, its explicit child references, transitive descendants, GitHub sub-issue children, and linked or closing PR evidence for those child issues. Use the same outbound traversal sources as A2, including closed umbrella children, so open descendants cannot be hidden behind a closed direct child. This step must not use repo-wide search to add unrelated work to the roadmap. The only repo-wide search allowed in A1.5 is a narrow duplicate/reuse check for a specific autonomous gap before creating a follow-up issue; use those results only to link existing gap work or avoid creating a duplicate, not to widen A2 candidates.
When the selected roadmap graph includes descendant issues that are
themselves roadmap nodes, such as descendants carrying the configured
roadmap label from labels.roadmapLabelName (default: roadmap) or a
idd-skill-roadmap-id marker, treat those
descendants as nested roadmaps rather than as normal execution
leaves. A nested roadmap is a coordination/audit node in the recursive
hierarchy: it may remain open while its own leaf descendants are still
executing, and its presence does not by itself widen A2 candidates
outside the selected roadmap graph.
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.
- 6d ago First seen · 191 lines · 2,405 tokens per session scan C 1479e0a984b0
idd-skill idd-roadmap-audit.instructions.md is an instructions file published in the GitHub repository kurone-kito/idd-skill (11 stars, last pushed today), licensed MIT. It adds 2,405 tokens to every session, about $0.0120 per session on Opus 5. A static security scan graded it C with 1 finding (hidden instructions). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other instructions, from other repositories
GRC-Mapping-Analyst GEMINI.md
Gemini CLI instructions for austinsonger/GRC-Mapping-Analyst, covering strm mapping — gemini cli context, purpose of this repository, when to perform strm mapping, working directory and saving completed mappings.
GRC-Mapping-Analyst AGENTS.md
AGENTS.md instructions for austinsonger/GRC-Mapping-Analyst, covering strm mapping — agent instructions, repository purpose, project constraints (apply to all work), activating the strm skill (codex) and strm methodology quick reference.
GRC-Mapping-Analyst copilot-instructions.md
Copilot instructions for austinsonger/GRC-Mapping-Analyst, covering github copilot instructions — strm mapping repository, core concepts, output files, file naming convention and strength score formula.
supergraph AGENTS.md
AGENTS.md instructions for datit309/supergraph, covering supergraph — mandatory workflows, skills, auto language detection, tiered workflow — pick the right tier first and full pipeline (tier 3).
mobile-ai-agents CLAUDE.md
Claude Code instructions for salmanashraf/mobile-ai-agents, covering claude.md — mobile ai agents, what this repo is, quick start, install everything and install by platform.
MemoryCustodian GEMINI.md
Gemini CLI instructions for waittim/MemoryCustodian, covering agent instructions, memorycustodian and development.