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 skills add mattgierhart/PRD-driven-context-engineering --skill ghm-template-syncgit clone --depth 1 https://github.com/mattgierhart/PRD-driven-context-engineeringWrote 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/skills/mattgierhart/prd-driven-context-engineering/ghm-template-sync)<a href="https://agentmods.dev/skills/mattgierhart/prd-driven-context-engineering/ghm-template-sync"><img src="https://agentmods.dev/badge/skills/mattgierhart/prd-driven-context-engineering/ghm-template-sync/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/skills/mattgierhart/prd-driven-context-engineering/ghm-template-sync"><img src="https://agentmods.dev/badge/skills/mattgierhart/prd-driven-context-engineering/ghm-template-sync.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- high Memory Poisoning · line 113 Skill manipulates agent memory, state, or stored context. Memory corruption can alter personality, override safety rules, or cause unpredictable behavior.Fix: Protect agent memory and state from modification by untrusted content. Use read-only memory for critical instructions and validate all state changes.
- medium Excessive Agency · line 76 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00082 | $0.01121 |
| Opus 5.5 | $0.00033 | $0.00448 |
| Sonnet 5.5 | $0.00016 | $0.00224 |
| Haiku 4.5 | $0.00008 | $0.00112 |
Grade A, and why
ghm-template-sync 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 — 119 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Template Sync
Detect template version drift and migrate to the latest template version.
Workflow
Phase 1: Detect Current State
- Read
.claude/VERSION(if it exists) to get current template version - If no VERSION file, check
CLAUDE.mdfrontmatter fortemplate_version - If neither exists, assume v1.0.0 (pre-versioning)
- Report: "Your repo is on template v{X}. Latest is v{Y}."
Phase 2: Diff Against Template
Compare your repo structure against what the current template version expects:
Check for missing files:
.claude/VERSION.claude/domain-profile.yaml.claude/hooks/HOOK_CONTRACT.md.claude/hooks/context-validation.sh.claude/hooks/context-density-gate.sh.claude/hooks/sot-update-trigger.shCHANGELOG.mdMIGRATION.md
Check for stale files (should be removed):
.claude/hooks/context-validation.py.claude/hooks/context-density-gate.py.claude/hooks/sot-update-trigger.py.claude/agents/HORIZON.md(replaced by subdirectory).claude/agents/STUDIO.md.claude/agents/DEVLAB.md.claude/agents/METRO.md
Check for structure issues:
settings.json: Does it use 3-level nesting? Are timeouts in seconds?- Agent directories: Do
horizon/,studio/,devlab/,metro/subdirectories exist withAGENT.md+MEMORY.md? - EPIC template: Does it use semantic headers (not numbered)?
- Frontmatter: Do key files have
template_version?
Phase 3: Generate Migration Plan
Output a table:
| File | Status | Action | Risk |
|---|---|---|---|
.claude/VERSION |
Missing | Copy from template | None |
.claude/hooks/*.py |
Stale | Delete (replaced by .sh) | None |
.claude/agents/HORIZON.md |
Stale | Split into horizon/AGENT.md + MEMORY.md | Preserve MEMORY.md content |
settings.json |
Outdated | Update hook nesting + commands | Check for custom hooks |
CLAUDE.md |
Missing frontmatter | Add template_version: "3.0.0" |
None |
Phase 4: Execute Safe Updates
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 · 119 lines · 82 tokens per session scan A 852eb2d1faf7
ghm-template-sync is a skill published in the GitHub repository mattgierhart/PRD-driven-context-engineering (181 stars, last pushed 1mo ago), licensed MIT. It adds 82 tokens to every session and 1,121 once invoked, about $0.0003 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 skills, from other repositories
remember
Review reusable project knowledge and decide what belongs in project memory, notepad, or durable docs.
blueprint-cloudflare-scaffold
Lay out a Worker with D1 and static Vue assets, with tests that run inside the Workers runtime.
project-design
Roteador de processos de design e marca que carrega orientações para criar arquivos DESIGN.md completos para um projeto e suas aplicações de interface.
self-assessment
Interactive skill assessment with personalized learning path generation.
loom-usage
Meta-orchestration skill for Claude driving loom itself.
loom-swift
Swift language expertise for idiomatic, production-quality code.