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 agentmods add skills/swingerman/engineer/prime-contextnpx skills add swingerman/engineer --skill prime-contextgit clone --depth 1 https://github.com/swingerman/engineerWhat 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 | $0.00048 | $0.00840 |
| Opus 5 | $0.00024 | $0.00420 |
| Sonnet 5 | $0.00010 | $0.00168 |
| Haiku 4.5 | $0.00005 | $0.00084 |
Grade A, and why
prime-context 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 2d 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 — 36 lines — stays where its author put it; the contents beside it link to each section on GitHub.
prime-context
Load working memory before pipeline work on a Ready feature — the convergent counterpart to discuss. Forked from superpowers:brainstorming but inverted: no exploration, just loading and orienting. Produces no artifact.
When to use
A prep step (not a checkpoint) between feature-init and discover-acs, on any feature the agent hasn't already worked this session.
Skip when: the agent just created the feature this session (context already warm). Not for: exploring whether a feature is worth doing (discuss).
Workflow
- Resolve + locate — resolve the methodology root + manifest via
${CLAUDE_PLUGIN_ROOT}/scripts/dae_resolve.py(seereferences/resolving.md); locate the feature (slug arg or branch name). Reject if no folder / nofeature.md. Layout disambiguation: if the project has BOTH afeatures/NNN-*/and aspecs/NNN-*/tree (a speckit migration in progress), state which holds this feature's artifacts; and if more than one acceptance-IR convention is present (features/NNN/.build/spec.jsonvs anacceptance-pipeline/ir/per-GWT layout), name the one this feature uses — so the load doesn't rediscover the split ad-hoc. (speckit-consolidateunifies it.) - Silent batch load — without narrating:
feature.md,CHARTER.md,manifest.yml, priorhandoffs/(especially the originating*-discuss.md), and the files named infeature.md's "Related code / design pointers". When loading those code pointers, prefer LSP —documentSymbolfor file shape,hoverfor signatures/types,workspaceSymbolto pull in the symbols they reference — over reading whole files blind, when an LSP MCP capability is available; fall back to grep + Read otherwise. See${CLAUDE_PLUGIN_ROOT}/references/code-lookup.md. - Orient — give a concise summary: outcome, scope, autonomy level (+ charter cap), key prior decisions, related code, relevant ADRs.
- One prompt — ask exactly one question: anything else to load? If the user names a new code pointer, load it and offer to add it to
feature.md. Then stop — prime-context orients, it does not interview. - Breadcrumb handoff — emit a tiny handoff recording what was loaded.
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.
- 2d ago First seen · 36 lines · 48 tokens per session scan A 6ea19461d581
prime-context is a skill published in the GitHub repository swingerman/engineer (144 stars, last pushed 6d ago), licensed MIT. It adds 48 tokens to every session and 840 once invoked, about $0.0002 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
writing-tests
Write unit tests, component tests, and integration tests for AiderDesk using Vitest and React Testing Library. Use when creating new tests, adding test coverage, configuring mocks, setting up test files, or debugging failing tests.
release
Cut a sim-use release end-to-end. Use when the user runs /release or asks to "ship a release", "publish a version", "cut a release", or "release to homebrew". Drives scripts/local-release.sh; never reimplement its build/sign/tarball logic.
work-summary
Generates a project dashboard from MCP work items. Default is a lean, attention-first view: what's in flight, what's blocked, what to do next, what's queued. Use when the user says: project status, what's active, show me the dashboard, work summary, what should I work on, project health, what's blocked, where did I…
manage-schemas
Creates, views, edits, deletes, and validates note schemas for the MCP Task Orchestrator in .taskorchestrator/config.yaml — the templates that define which notes agents must fill at each workflow phase. Also recommends schema designs from a library of workflow patterns (autonomous coding loops, spec-driven teams…
bmad-dev-story
Execute story implementation following a context filled story spec file. Use when the user says "dev this story [story file]" or "implement the next story in the sprint plan".
bmad-editorial-review-structure
Structural editor that proposes cuts, reorganization, and simplification while preserving comprehension. Use when user requests structural review or editorial review of structure.