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 conn-castle/agent-layer --skill instruction-syncgit clone --depth 1 https://github.com/conn-castle/agent-layerWrote 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/conn-castle/agent-layer/instruction-sync)<a href="https://agentmods.dev/skills/conn-castle/agent-layer/instruction-sync"><img src="https://agentmods.dev/badge/skills/conn-castle/agent-layer/instruction-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/conn-castle/agent-layer/instruction-sync"><img src="https://agentmods.dev/badge/skills/conn-castle/agent-layer/instruction-sync.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 | $0.00036 | $0.00488 |
| Opus 5.5 | $0.00014 | $0.00195 |
| Sonnet 5.5 | $0.00007 | $0.00098 |
| Haiku 4.5 | $0.00004 | $0.00049 |
Grade C, and why
instruction-sync 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 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.
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.
<!-- agent-layer-catalog-skill: instruction-sync --> What it actually says
Instruction sync
Read al instructions --help and the relevant subcommand help before operating.
Start with al instructions status --all, which is offline. al sync projects
current local content without fetching. Use pull to update imports.
al instructions add <repository> <exact/path.md> --order <integer> --yes
imports one regular Markdown file. Order is required, nonnegative, globally
unique across imported and local blocks; zero and gaps are valid. Source ref,
tracking, write policy, push repository and push branch use the skills defaults
and flags. Order is configuration-owned and never published upstream.
diff <filename.md> accepts --from and --to base, local, upstream, or
destination. pull preserves edits through a three-way merge. For a conflict,
resolve the reported Git workspace index, then run resolve <filename.md>.
reset <filename.md> --yes discards local imported edits. remove <repository> <selector> --yes removes one import block; modified files block retirement.
push --yes respects configured write policies and excludes project-local files.
Use destructive commands and upstream writes only when the user authorizes them. Import only known repositories/files; do not search or recommend sources.
Project-owned files stay in .agent-layer/instructions/, each declared once in
[[instructions.local]] with selectors = ["filename.md"] and order = N.
Imports live in .agent-layer/instructions-imported/ with shared lifecycle state
in instructions.lock.json. Edit imported files there, then sync.
Legacy projects need al upgrade or accepted wizard changes for offline ordering.
The wizard can adopt existing standard rules/memory through an explicit networked
import, preserving their bytes, permissions and order atomically. None leaves
existing content unchanged. Unsafe linked instruction sources must be made local
before migration. Do not replace these operations with manual lock edits.
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 · 42 lines · 36 tokens per session scan C 1bcc56f1450b
instruction-sync is a skill published in the GitHub repository conn-castle/agent-layer (10 stars, last pushed today), licensed MIT. It adds 36 tokens to every session and 488 once invoked, about $0.0001 per session on Opus 5.5. 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 2026-10-12.
Other skills, from other repositories
moai-ref-git-workflow
Git workflow patterns, branch strategies, conventional commits, and PR templates reference for git operations. Agent-extending skill that amplifies manager-git expertise with production-grade git workflow patterns. NOT for: code implementation, testing, architecture design, documentation content.
commit
Create git commits with good messages. Use when user says "commit", "create commit", or asks to commit changes.
release-notes
Draft a release note section from a commit range, grouped by what the reader gets.
kirocrew-prepare-pr
LOAD THIS FOR EVERY KIRO CREW PR you open or update (Kiro Crew repo only; not other repos or CRs). Runs the whole loop — issue + tier, commit, sync, squash, push, drive CI and AI review to green, fix a red main, arm auto-merge and watch it land (24h). 'push my changes' = prepare-only.
commit
Atomic git commit with conventional message. Use when the user says "commit", "save my changes", "commit this", or wants to create a git commit. Stages specific files, writes a conventional commit message with body explaining non-obvious decisions. Never uses git add -A.
akigitcommit
Analyze the working tree and commit changes in clean logical groups. Triages a long half-finished tree first (finished vs mid-edit vs abandoned vs accidental) before grouping. Auto-detects CHANGELOG to switch between domain-grouped mode (3–5 commits by object/feature) and type-grouped mode (feat/fix/refactor). Stages…