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 agagniere/speky --skill speky-workflowgit clone --depth 1 https://github.com/agagniere/spekyWrote 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/agagniere/speky/speky-workflow)<a href="https://agentmods.dev/skills/agagniere/speky/speky-workflow"><img src="https://agentmods.dev/badge/skills/agagniere/speky/speky-workflow/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/agagniere/speky/speky-workflow"><img src="https://agentmods.dev/badge/skills/agagniere/speky/speky-workflow.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 | $0.00022 | $0.01248 |
| Opus 5 | $0.00011 | $0.00624 |
| Sonnet 5 | $0.00004 | $0.00250 |
| Haiku 4.5 | $0.00002 | $0.00125 |
Grade A, and why
speky-workflow 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 8d 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 — 84 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Speky Workflow Guidelines
Speky organizes a project's detailed specification into requirements (what the system must do) and tests (how to verify it). Items are identified by short IDs like RF001, RN02, T003.
Tool selection
- Start with
list_all_idsto get a full overview of available IDs before searching or fetching. - Use
get_requirement/get_testwhen you already know the ID — it's faster and returns full detail including cross-references. - Use
search_requirementsonly when filtering by tag or category. Usesearch_testsonly when filtering by category or by requirement (tester_of). Do not call either with no filters as a substitute forlist_all_ids. - Use
list_all_tagsto discover available tags before filtering by one. - Use
list_references_toto find what depends on a given item (e.g., before modifying or deleting it). - Use
test_plan_coverageto find coverage gaps — it gives a project-wide summary without requiring manual inspection. - Use
least_tested_requirementsto find where to write new tests — it ranks requirements by ascending test count and acceptstag,category, andcountfilters.
Reading results
get_requirementreturnstested_by(tests covering it) andreferenced_by(requirements that depend on it). Read these before assessing coverage or impact.get_testreturnsref(requirements it validates) andprereq(tests that must run first). Read these to understand test scope and ordering.- Tags use a namespaced format:
namespace:value(e.g.,output:pdf,mcp:tools). When filtering by tag, use the exact full string.
Working with coverage
- A requirement is covered if it appears in at least one test's
reflist (visible astested_byon the requirement). - Use
test_plan_coveragefor a project-wide summary before concluding anything about overall coverage. - Use
least_tested_requirements(with optionaltagorcategoryfilter) to pinpoint the requirements most in need of new tests. - Do not infer coverage from description alone — always check
tested_by.
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.
- 8d ago First seen · 84 lines · 22 tokens per session scan A 4af4a1ca49b4
speky-workflow is a skill published in the GitHub repository agagniere/speky (2 stars, last pushed 3mo ago), licensed MIT. It adds 22 tokens to every session and 1,248 once invoked, about $0.0001 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 2026-08-31.
Other skills, from other repositories
runtime-observation
Cross-cutting skill for runtime observation methodology. Five-phase observation discipline, container interaction patterns, observation record format, claim extraction, environment variation. Loaded by the analyzer agent for runtime observation roles.
test-suite-analysis
Layer 1 skill for extracting behavioral intelligence from test suites. Framework detection, test code reading strategy, test execution strategy, behavioral claim extraction with Given/When/Then mapping, e2e vs unit value classification. Loaded by the analyzer agent during Layer 1.
validation-methodology
Cross-cutting validation discipline. Acceptance criteria format, definition of done checklists, quality gate criteria, verification methods. Loaded by every analysis agent.
fidelity-validation
Cross-validates sanitized output specs against raw source specs to detect lost behavioral detail, dropped constants, missing features, or diluted precision. Run AFTER sanitization and AFTER contamination audit passes.
source-completeness
Validates that all user-facing surfaces discovered in source code are captured in the behavioral specs. Catches coverage gaps where the analysis pipeline missed features, tools, env vars, CLI flags, or other behavioral interfaces. Run AFTER Layer 3 specs are written, BEFORE sanitization.
live-preview
Mid-build visual verification loop. Takes screenshots of components during construction, not just after. Catches visual regressions and invisible features before they compound. Requires Playwright or similar screenshot tool.