Borrowing it
Nothing to install: this file belongs to sirkirby/unifi-mcp. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/sirkirby/unifi-mcp/main/.agents/skills/skill-reference-and-manifest-drift-maintenance/SKILL.mdgit clone --depth 1 https://github.com/sirkirby/unifi-mcpWrote 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/sirkirby/unifi-mcp/skill-reference-and-manifest-drift-maintenance)<a href="https://agentmods.dev/skills/sirkirby/unifi-mcp/skill-reference-and-manifest-drift-maintenance"><img src="https://agentmods.dev/badge/skills/sirkirby/unifi-mcp/skill-reference-and-manifest-drift-maintenance/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/sirkirby/unifi-mcp/skill-reference-and-manifest-drift-maintenance"><img src="https://agentmods.dev/badge/skills/sirkirby/unifi-mcp/skill-reference-and-manifest-drift-maintenance.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, 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 Prompt Injection · line 31 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
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 | $0.00224 | $0.03034 |
| Opus 5 | $0.00112 | $0.01517 |
| Sonnet 5 | $0.00045 | $0.00607 |
| Haiku 4.5 | $0.00022 | $0.00303 |
Grade A, and why
myco:skill-reference-and-manifest-drift-maintenance 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 5d 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 — 227 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill Reference & Generated Manifest Drift Maintenance
Each app has a generated tool manifest (e.g.
apps/protect/src/unifi_protect_mcp/tools_manifest.json) and a generated
skill reference doc under the corresponding plugin (e.g.
plugins/unifi-network/skills/unifi-network/references/network-tools.md,
plugins/unifi-access/skills/unifi-access/references/access-tools.md)
that uses <!-- AUTO:tools:<category> --> marker blocks. Both are
generated artifacts — a pure function of the Python source
(docstrings, parameter signatures, tool registrations).
packages/unifi-mcp-shared/src/unifi_mcp_shared/manifest_generator.py
implements the shared manifest-generation logic that each app's own
generator wrapper script calls (e.g.
apps/network/scripts/generate_tool_manifest.py, invoked via that app's
make manifest target). Nothing about these generated files should ever
be hand-authored. The recurring failure mode in this codebase is not that
regeneration is hard — it's that CI's drift-check only diffs
already-rendered content against disk. It cannot detect a missing
section, a missing marker, or a description that's technically present
but shallower/staler than the source. Those gaps are only caught by a
human applying the checklists below.
Prerequisites
- Locate the generator scripts for this repo before starting — search
scripts/(e.g.Grep -r "tools_manifest" scripts/andGrep -r "generate_tool_manifest\|generate_skill_references" scripts/) rather than assuming a fixed filename, since generator script names can move between refactors. Verified examples in this repo:scripts/generate_skill_references.py, the sharedpackages/unifi-mcp-shared/src/unifi_mcp_shared/manifest_generator.pymodule, and the per-app wrapper it documents (e.g.apps/network/scripts/generate_tool_manifest.py, run via that app'smake manifesttarget). - Locate the skill reference doc(s) for the category you're touching by
searching repo-wide for
AUTO:tools:markers (Grep -rl "AUTO:tools:" .) rather than assuming a fixed path. These live under each plugin's own skill references directory (plugins/<plugin>/skills/<plugin>/references/*.md) — not underapps/— and the exact category set can vary per plugin. - Regenerate inside a worktree venv synced to the locked dependencies. Using system Python or a stale venv can produce output that's subtly different from what CI would generate, defeating the point of regenerating at all.
make pre-commitdoes regenerate manifests when it's actually run — both the root Makefile (pre-commit: format generate lint test check-generated worker-typecheck, wheregenerateincludesmanifest) and each app's own Makefile (e.g.apps/network/Makefile:pre-commit: format manifest server-manifest lint test). But there is no.pre-commit-config.yamlwiring it into a git hook — it's a manualmaketarget. Don't assume a branch's manifest is current just because its author "probably ran pre-commit"; check explicitly (Procedure A) whenever behavior-relevant source changed.
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.
- 5d ago Changed · +12 lines 4d607f456a53
- 13d ago First seen · 215 lines · 224 tokens per session scan A 3479a8e4635c
myco:skill-reference-and-manifest-drift-maintenance is a skill published in the GitHub repository sirkirby/unifi-mcp (814 stars, last pushed 2d ago), licensed MIT. It adds 224 tokens to every session and 3,034 once invoked, about $0.0011 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 skills, from other repositories
streamdeck-designer
Design, theme, and author complete Stream Deck layouts for the user's hardware using the streamdeck-mcp tools. Use whenever the user wants a custom Stream Deck setup, asks for a themed deck (e.g. "hello-kitty Twitch deck", "retrowave music controls", "a dev deck for this repo"), wants buttons or dials configured for…
streamdeck-profile
Use this skill when you need to configure Elgato Stream Deck desktop profiles without taking USB control of the hardware.
unifi-udm-pro-mcp
UniFi UDM Pro MCP Server for managing networks, connected clients, Wi-Fi settings, and device health via Model Context Protocol.
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
specflow-use
To connect Rosetta with Grid Dynamics SpecFlow MCP; only when SpecFlow is mentioned and the MCP is installed.
ue-mcp-epic-routing
Use when deciding between ue-mcp's native category actions and Epic's wrapped ToolsetRegistry tools (the epic actions, incl. the Blueprint graph DSL) for a task in Unreal. Pulls in when authoring Blueprint graph bodies, or any time both a native action and an epic action could do the job and you need to pick.