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 randomittin/heimdall --skill stacksgit clone --depth 1 https://github.com/randomittin/heimdallWrote 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/randomittin/heimdall/stacks)<a href="https://agentmods.dev/skills/randomittin/heimdall/stacks"><img src="https://agentmods.dev/badge/skills/randomittin/heimdall/stacks.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 | $0.00085 | $0.00748 |
| Opus 5 | $0.00043 | $0.00374 |
| Sonnet 5 | $0.00017 | $0.00150 |
| Haiku 4.5 | $0.00009 | $0.00075 |
Grade A, and why
stacks 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 yesterday.
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 — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.
stacks
Knowledge packs loaded onto existing role agents — never new role×stack agents. There is
exactly one coder, one reviewer, one architect; a pack gives that same agent
stack-specific judgment without changing its identity. Full design: skills/stacks/README.md.
Run it
bin/stack-detect [path] # scan manifests, print detected stack id(s) as JSON
bin/stack-pack load [path] # detect + print the pack file path(s) to read, in order
bin/stack-pack load prints, in layering order:
- Base pack —
skills/stacks/<id>/PACK.md(cold-start scaffold, shipped with hmd). - Repo refinement — any
.planning/skills/*.mdin the target project (learned, repo-specific notes that override the base pack where they disagree).
Read every path it prints, then apply those conventions for the rest of the task — exact lint/format/test/build commands, directory layout, runnable acceptance-criteria templates, common failure patterns + fixes. Your identity as the current role does not change; only your knowledge of this stack does.
Detection already runs for you
A SessionStart hook runs bin/stack-detect and writes the result to
.planning/detected-stack.json when the project is a Heimdall project. If that file
exists, read it directly instead of re-running detection, then call bin/stack-pack load
(or resolve the paths yourself from the ids it lists) to get the pack(s) to read.
Supported stack ids
nextjs · react-native · react · vue · svelte · node · spring-boot · jvm ·
fastapi · django · flask · python · go · rust — detection rules for each:
skills/stacks/README.md.
Adding a new pack
Copy STACK_PACK_TEMPLATE.md (repo root) to skills/stacks/<stack-id>/PACK.md, fill every
section, and confirm bin/stack-detect already emits that <stack-id> — a pack whose id
stack-detect never emits is never loaded.
Verification
-
bin/stack-detect .on a repo whosepackage.jsondepends onnextprints{"stacks":["nextjs"], ...}(asignalsarray included). -
bin/stack-pack load .on that same repo prints a path ending inskills/stacks/nextjs/PACK.md. - A repo that also has a
.planning/skills/*.mdfile gets that path printed AFTER the base pack path — repo refinement layers on top, never instead of. - A repo matching none of the known stacks:
bin/stack-detect .prints{"stacks":[],"signals":[]}andbin/stack-pack load .prints no pack path.
What ships with it
5 files 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.
- yesterday First seen · 58 lines · 85 tokens per session scan A cfe7455a41b0
stacks is a skill published in the GitHub repository randomittin/heimdall (5 stars, last pushed today), licensed MIT. It adds 85 tokens to every session and 748 once invoked, about $0.0004 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 2026-09-05.
Other skills, from other repositories
memstack-automation-n8n-workflow-builder
Use this skill when the user says 'n8n workflow', 'build a workflow', 'automation workflow', 'connect services', or needs visual workflow design with node mapping, data transformations, and error handling for n8n. Do NOT use for standalone webhook endpoints or cron jobs.
atomic-writing
One voice for every file the repo ships: README, docs/guides/, docs/reference/, docs/spec/, docs/design/, docs/research/, docs/wiki/, CLAUDE.md, and the prompt artifacts under commands/, agents/, skills/, rules/, and output-styles/. Clear, direct, technical, and visual wherever the content has a shape. No marketing…
memstack-automation-cron-scheduler
Use this skill when the user says 'cron job', 'scheduled task', 'run every', 'cron expression', 'recurring job', or needs production-grade scheduled jobs with overlap prevention, monitoring, and structured logging. Do NOT use for n8n workflows or event-driven webhooks.
atomic-documentation
Diff-driven documentation surface classifier. Given a diff (staged, branch, or range), reads the project's indexed ## Documentation surfaces table from CLAUDE instructions, matches the diff against it, and emits a structured list of proposed edits. Two modes: maintenance (commit flow — stale/incomplete only, never…
atomic-visual-options
Planning-phase visual comparison aid. Renders 2-4 side-by-side variants per decision dimension as a single throwaway, self-contained HTML file and captures the user's pick as typed terminal codes (e.g. "A2 B3"). Auto-fires on phrases like "show me a few options", "mock up some variants", "let me see this side by…
atomic-git-discipline
Compressed commit message and PR body generator. Cuts noise from both while preserving intent and reasoning. Conventional Commits format. Subject ≤50 chars, body only when "why" isn't obvious; PR bodies state only what the diff can't show, 120 words max. Use when user says "write a commit", "commit message", "generate…