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/hulupeep/specflow/specflow-simulatenpx skills add Hulupeep/Specflow --skill specflow-simulategit clone --depth 1 https://github.com/Hulupeep/SpecflowWrote 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/hulupeep/specflow/specflow-simulate)<a href="https://agentmods.dev/skills/hulupeep/specflow/specflow-simulate"><img src="https://agentmods.dev/badge/skills/hulupeep/specflow/specflow-simulate.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 | $0.00172 | $0.00940 |
| Opus 5 | $0.00086 | $0.00470 |
| Sonnet 5 | $0.00034 | $0.00188 |
| Haiku 4.5 | $0.00017 | $0.00094 |
Grade A, and why
specflow-simulate 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 4d 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 — 43 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Specflow Simulate
Run a story through realistic usage — multiple personas, multiple routes — to find the gaps and edges a single happy-path reading misses. The output is not commentary; it is a set of concrete, proposed additions that make the story sharper before any code is written.
When to run
Trigger on: "simulate this story", "simulate usage end to end", "find gaps / edges", "run personas through this", "stress-test this story", or /specflow-simulate <file-or-issue>.
Position in the Specflow flow — this runs after create, before audit:
create (specflow-writer) → SIMULATE (this skill) → audit/uplift (specflow-audit) → pre-flight gate
It is generative (discover + propose). It is NOT the pre-flight gate (that verifies and blocks). Do not conflate the two.
The method
Full method (persona archetypes, route taxonomy, gap→edit mapping, output format) is in references/simulation-method.md. Summary:
- Read the story (issue via
gh issue view <n> --json title,body,comments, or the file). Extract the intended outcome, the actors, and the surfaces. - Derive 3–5 distinct personas — not demographics; behavioural archetypes that route through the problem differently (e.g. power user, first-timer, adversarial/edge user, wrong-permissions user, interrupted/returning user). Pick the ones that stress this story.
- Walk each persona through the story end-to-end — the happy path AND at least one divergent route per persona: where do they get confused, blocked, take an unanticipated branch, hit an empty/error/permission state, or make a decision the story didn't plan for?
- Collect gaps — each gap is a concrete moment where the story is silent, ambiguous, or wrong for that persona/route.
- Map every gap to a story edit — a new
REQ-NN, a negative-pathAC, a new Gherkin scenario/branch, a new journey step +data-testid, or a flagged open question. A gap with no proposed edit is incomplete. - Propose, don't silently rewrite — post the findings as ONE labelled block: a GitHub comment (
gh issue comment) or an appended## Simulation Findings (proposed)section. The user (or the subsequent specflow-audit uplift) folds accepted items into the story body. Only edit the body directly if the user explicitly says to apply.
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.
- 4d ago First seen · 43 lines · 172 tokens per session scan A ef0033b37ab5
specflow-simulate is a skill published in the GitHub repository Hulupeep/Specflow (24 stars, last pushed 1mo ago), licensed MIT. It adds 172 tokens to every session and 940 once invoked, about $0.0009 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
truecourse-hooks
Install, configure, or remove the TrueCourse pre-commit hook.
truecourse-analyze
Run TrueCourse architecture analysis on this repository.
truecourse-fix
Fix TrueCourse violations that have suggested fixes.
truecourse-list
List TrueCourse violations found in this repository.
code-quality-review
Run a maintainability and structure review focused on abstraction quality, branching complexity, file growth, canonical ownership, duplication, and refactoring opportunities. Use when the user asks for code quality review, maintainability review, 代码质量审查, 可维护性审查, or comments about whether the change stays easy to…
optimization-audit
Comprehensive optimization audit with two modes and a single tier. Planning mode designs performance strategy, capacity planning, and scaling architecture. Audit mode scans code and infrastructure for performance anti-patterns, inefficient algorithms, N+1 queries, missing caching, concurrency issues, and resource…