jira-task — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited jira-task (Agent Skill) and scored it 92/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 2 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 2 flagged
The text {match} tells the agent to skip the normal "ask the user first" gate. Used adversarially it removes the human-in-the-loop check before destructive or sensitive actions, turning a normally-gated agent into a fire-and-forget executor.
A bulleted imperative like {match} tells the agent to never reveal, disclose, or mention something to the user. Used adversarially it can instruct the agent to hide its tool calls or lie about what it did — stripping the transparency a user relies on to trust the agent.
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Use this skill only to generate Jira Task ticket content. A Task describes bounded work with an expected outcome and verifiable completion criteria.
This skill outputs Jira Markup in chat only. It does not create Jira issues through the Jira API.
Assumptions and TBD only for real unresolved non-blocking gaps.? or using Spanish ¿...? when the conversation language is Spanish.Focused Task question.Use Quick Creation unless the user explicitly asks for refinement. Load assets/quick-creation-template.md before generating.
Generate directly when the user provided enough information for:
Ask only one blocking question if any of those three cannot be inferred safely. Keep the lead-in minimal before the Jira block.
Use Refinement only when the user explicitly asks to refine, improve, complete, add detail, clarify, make developer-ready, add acceptance criteria, or equivalent Spanish intent such as refinar, mejorar, completar, agregar detalle, hacer mas claro, dejar listo para dev, or anadir criterios de aceptacion.
Use SDD-ready Refinement when the user explicitly asks for listo para SDD, preparalo para SDD, para usarlo en SDD, SDD-ready, ready for SDD, prepare for SDD, as SDD input, or equivalent phrasing.
Load assets/refinement-template.md before generating. Ask one question at a time using this structure:
### Question N — [direct interrogative question?]
**Recommended answer:** Provide a short suggested answer.
**Why this matters:** Explain why this answer affects the Task.
**Estimated remaining questions:** ~MDo not generate until the user explicitly asks to generate, create, print, or finalize the ticket.
Required refined sections:
Optional refined sections:
Acceptance Criteria should use Given/When/Then by default. Verifiable bullets are allowed for pure technical work when Given/When/Then would be artificial.
Implementation Notes may include technical guidance, constraints, dependencies, risks, known technical context, or SDD inputs. They must not become a step-by-step implementation plan, invented architecture, or subtasks disguised as design.
Open Questions are especially useful for SDD-ready output when unresolved decisions affect scope, design, or acceptance.
When required, place this before the Jira block:
Evidence Summary Reviewed: Files, docs, or sources reviewed. Found: Facts discovered. Evidence: Specific references. Confidence: High | Medium | Low Unverified: Claims or assumptions not validated.
If no exploration happened and no technical or factual claims are made, omit the Evidence Summary.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.