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 sjarmak/coding-agent-workflows --skill project-compassgit clone --depth 1 https://github.com/sjarmak/coding-agent-workflowsWrote 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/sjarmak/coding-agent-workflows/project-compass)<a href="https://agentmods.dev/skills/sjarmak/coding-agent-workflows/project-compass"><img src="https://agentmods.dev/badge/skills/sjarmak/coding-agent-workflows/project-compass/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/sjarmak/coding-agent-workflows/project-compass"><img src="https://agentmods.dev/badge/skills/sjarmak/coding-agent-workflows/project-compass.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 | $0.00069 | $0.01618 |
| Opus 5.5 | $0.00028 | $0.00647 |
| Sonnet 5.5 | $0.00014 | $0.00324 |
| Haiku 4.5 | $0.00007 | $0.00162 |
Grade A, and why
project-compass 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 — 136 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Project Compass
Map the tribal knowledge of a codebase into per-area COMPASS.md files: the
why and the gotchas, the things that aren't visible in the code's surface and
that otherwise live only in someone's head. One map per major area, living next
to the code it describes. Inspired by mapping tribal knowledge across large
codebases (per Meta's data-pipeline knowledge-mapping work).
Compass files are the dense layer. They exist so that AGENTS.md and memory
stay thin — see the context-layering practice for the boundary. AGENTS.md only
indexes compass files; it never copies their content.
When to use
- "Map this codebase" / "generate compass files" / "set up codebase compass"
- "Refresh the compass for
src/api" after that area changed materially - During
project-init, to seed maps for the largest areas - The user asks "why is this area shaped this way?" (read the existing compass first)
ZFC boundary
This skill is AI-orchestration. Mechanism gathers raw signal; the model writes every word of the map. Do not template-fill prose from heuristics — a compass is a judgment about what matters in an area, which is exactly what a model does well and a regex does not.
- Mechanism: file discovery, directory sizing, content-hashing, reading
source, writing the file, updating the index. The hash is computed by
compass-hash.mjs(ships in this skill dir), never by the model. - Model: purpose, what counts as a key entry point, which constraints are non-obvious, which gotchas are load-bearing.
How it works
Phase 1 — Pick the areas (mechanism)
Identify the major areas worth a map. Reuse codebase-onboarding reconnaissance
for structure detection. An "area" is a directory that is both substantial
(meaningful share of the code) and cohesive (one responsibility). Skip
vendored, generated, and build directories. For a small repo, one root-level
compass may be enough; for a monorepo, one per top-level package.
Confirm the candidate area list with the user before generating — generating a map per leaf directory is the failure mode to avoid.
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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 · 136 lines · 69 tokens per session scan A 540ec2ad2346
project-compass is a skill published in the GitHub repository sjarmak/coding-agent-workflows (6 stars, last pushed 17d ago), licensed MIT. It adds 69 tokens to every session and 1,618 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 2026-08-31.
Other skills, from other repositories
planning-with-files
Persistent file-based planning for multi-step AI-agent work. Keeps taskplan.md, findings.md, and progress.md on disk; lifecycle hooks inject selected project planning context. Automatic recovery reads project planning files only. Explicit session-catchup.py --metadata reads same-project local agent session records and…
ctx-purge
Purge the context-mode knowledge base. Permanently deletes all indexed content and resets session stats. This is destructive and cannot be undone. Trigger: /context-mode:ctx-purge.
forgetful-procedures
Store and find procedural knowledge — reusable step-by-step how-tos — in Forgetful's skill store. Use when a workflow worth repeating gets worked out, when the user asks how something is done here, or when importing or exporting SKILL.md files. Litmus: a procedure someone follows is a skill; a fact someone recalls is…
memory-md-management
Provides comprehensive memory file management capabilities including auditing, quality assessment, and targeted improvements for files such as CLAUDE.md. Use when user asks to check, audit, update, improve, fix, maintain, or validate project memory files. Also triggers for "project memory optimization", "CLAUDE.md…
context-window-engineer
Expert guide for engineering and optimizing ultra-large context windows (2M+ tokens) across Gemini 4 Pro, Claude 5.5, and GPT Astra 6 — context partitioning, retrieval-augmented context injection, sliding window strategies, and cost-aware token budgeting / Panduan ahli rekayasa dan optimasi context window ultra-besar…
context-degradation
Recognize patterns of context failure: lost-in-middle, poisoning, distraction, and clash.