Menu
SkillsMP has collected 257 skills from cyberskill-official/cyberos. Open a skill to review its source and details.
Showing 40 of 257 collected skills.
Audit a hardening-record@1 for scope discipline, verification honesty, and gate integrity. Checks every changed file was inside a worked finding's declared scope, every closed finding carries verbatim validation output and a regression gate, fingerprints were…
Work findings from an inspection-report@1 under scope and safety discipline and emit a hardening-record@1. Validates input with inspect-lint before reading findings, classifies each finding as agent/operator/split (honouring operator_prerequisites), halts at…
Run the CyberOS ship-tasks workflow in any repo: drive a task (product or improvement class) through implement -> review -> test -> done, with HITL required at the two human-acceptance gates and gates from the repo's own build/lint/test. Use when asked to…
Audit an existing inspection-report@1 against inspection_report_rubric@1.0. Runs tools/inspect-lint.mjs as a machine floor, then scores judgement rules a linter cannot check. Emits a score out of 10 and refuses to pass below 10/10. Use when the user asks to…
Author a full read-only project inspection (inspection-report@1) across 75 engineering disciplines (spec ≥1.2; 69 rows remain valid for INSPECT-SPEC 1.0 goldens via version gating), producing a 22-section report with a coverage ledger, evidence-graded…
Spec correctness gate — audit one or more existing task@1 markdowns against audit_rubric@2.0 (FM/SEC/COND/QA/SAFE/TRACE rule families) to drive the `draft → ready_to_implement` lifecycle transition per `modules/skill/contracts/task/STATUS-REFERENCE.md` §1.1.…
Generate a versioned, audited Task backlog from one or more PRD/spec/SRS documents. Use when user asks to "draft a task", "turn this PRD into a backlog", or "expand this spec into tasks". Halts at PLAN approval, HITL gates, and amendment batches; resumable…
Audit an architectural-spike@1 against architectural_spike_rubric@1.0: enforces structural completeness (SPK-STRUCT - frontmatter, five sections, recommendation names exactly one probed option), checkable evidence on every option (SPK-EVID - repo path,…
Run a time-boxed architectural spike when an ADR is pending with >= 2 viable options, or the task introduces a dependency the repo has never used - producing architectural-spike@1: the single question under investigation, options probed with CHECKABLE…
Audit a backlog-state-update@2 (accepting @1 for one release window) against backlog_state_update_rubric@2.0: enforces new_status in the 10-value enum, line_number/old_line optimistic concurrency for status-cell mutations, the BSU-INS family for insert-row…
Write `docs/tasks/BACKLOG.md` mutations as backlog-state-update@2: rewrite one status cell atomically (used by `chief-technology-officer/ship-tasks` between every phase; the same write emits the `workflow_complete` memory row), or INSERT one new row…
Test coverage gate audit (testing → done) — audit a coverage-gate@1 artefact against coverage_gate_rubric@1.0: enforces tests_failed==0, files_below_90pct empty, ecm_rows_uncovered empty, raw_terminal present + non-truncated, AND every §1 clause's cited test…
Test coverage gate (testing → done) — run the project's test suite + measure coverage on the files touched by the current task (per the `git diff` since the task's `implementing` status was set). Emits a coverage-gate@1 artefact: raw terminal output of the…
Audit a debug-trace@1 against debugging_cycle_rubric@1.0: enforces budget compliance, non-vacuous hypotheses, resolvable file:line references, correct circuit-breaker arithmetic, and a defined resolution. Emits a `score / 10` verdict; refuses to pass on…
Run a multi-vector debugging pass when the coverage-gate trips (tests_failed > 0 OR files_below_90pct non-empty OR ecm_rows_uncovered non-empty). For each failed iteration: (1) classify the failure vector — state / network / memory / logic / flake; (2) state…
Audit an edge-case-matrix@1 against edge_case_matrix_rubric@1.0: enforces ≥1 row per category, SECURITY rows pointing at real test paths, DEGRADATION rows specifying detection + recovery, and `total_rows ≥ 8` for MUST-priority tasks. Emits a `score / 10`…
Generate a structured edge-case-matrix@1 for a task before implementation. Enumerates: null/empty inputs, extreme bounds (off-by-one, integer overflow, time-zone DST, leap second, Unicode normalisation), malformed payloads (truncated, oversized, non-UTF-8,…
Audit a mock-contract-test@1 against mock_contract_test_rubric@1.0: enforces ≥1 request_response_pair, error_modes coverage of every SECURITY/DEGRADATION matrix row, swap_target is a real symbol, sunset_criterion has an observable trigger, and contract_tests…
When a task declares an external dependency that does not yet exist (missing API key, future service, paywall, 2FA challenge, CAPTCHA, third-party that needs procurement), author a `mock-contract-test@1` artefact: (a) the exact expected Request/Response shape…
Audit an observability-injection@1 against observability_injection_rubric@1.0: enforces ≥1 log_point per state transition, ≥1 trace_span per external IO, ≥1 error_counter per error branch, branch_coverage ≥80%, redaction_policy present when PII is in scope.…
Walk the critical paths of the task's implementation plan and emit an `observability-injection@1`: (a) structured-log lines at every state transition (always carrying tenant_id + subject_id when in-scope), (b) trace spans wrapping every external IO call, (c)…
Audit a `plan@1` against `plan_rubric@1.0` and refuse to pass below 10/10. REDS a plan that is missing an option (PLAN-OPT-001, fewer than two options weighed), missing a decision (PLAN-DEC-001, zero or more than one decision), or missing the out list…
Turn an idea into a `plan@1` that create-tasks can consume unmodified - the front door create-tasks promised but never had (its standalone interview asks for a source file, not an idea). Detects mode from the inputs: greenfield (an idea, no repo), brownfield…
Audit a repo-context-map@1 against repo_context_map_rubric@1.0: enforces presence of the three baseline patterns (error_type, logging, test_framework), `pinned_in` references that resolve to real files, schemas present when the task declares migrations, and…
Deep-scan the repo before any code is generated for a given task, and emit a `repo-context-map@1` capturing: (a) existing patterns the new code must follow (DI containers, error type, state-management style, logging convention), (b) database schemas + type…
Evidence ladder for a task whose status claims work ship-tasks did not perform — already implemented mid-shipping or long "done", with no run manifest, a manifest that fails verify, or a missing phase artefact set. Runs `docs-tools/task-reconcile.mjs` as its…
The outer loop: reads a bounded window of the workflow's own run exhaust — gate logs, route-back reasons, reconcile reports (NOT memory rows - see below) reports — clusters the shapes that RECUR, and proposes at most three `skill-amendment@1` records naming…
Audit one or more existing account-plan@1 markdowns against account-plan_rubric@1.0. Produces a sibling .audit.md per artefact plus an AUDIT_BATCH_SUMMARY. Halts on needs_human verdicts; resumable on audited_file_sha256. Standalone trigger or chains naturally…
Author a per-strategic-account plan (context + org chart + current spend + expansion thesis + threats + quarterly action plan + executive sponsor coverage). CRO-Revenue or CCO-Customer authored. Use when user asks to "draft a account plan" or "create the…
Audits an ai-strategy@1 artefact against ai_strategy_rubric@1.0. Validates FM + SEC structure, EU AI Act risk-classification per use case, NIST AI RMF coverage, MLOps maturity self-assessment honesty, ethics-posture-vs-use-cases internal consistency. Per NIST…
Authors the annual AI strategy — use-case portfolio (business value × feasibility × risk), build/buy/partner per use case, model governance posture, training-data strategy, MLOps maturity (Google MLOps levels), EU AI Act risk classification per use case…
Audit one or more existing ai-use-case-portfolio@1 markdowns against ai-use-case-portfolio_rubric@1.0. Produces a sibling .audit.md per artefact plus an AUDIT_BATCH_SUMMARY. Halts on needs_human verdicts; resumable on audited_file_sha256. Standalone trigger…
Author the quarterly AI use-case portfolio review (executive summary + use-case inventory by stage + EU AI Act risk-class distribution + production-deployment count + ROI tracking + kill-list candidates + pipeline coverage). CAIO authored. Use when user asks…
Audit one or more existing analyst-briefing@1 markdowns against analyst-briefing_rubric@1.0. Produces a sibling .audit.md per artefact plus an AUDIT_BATCH_SUMMARY. Halts on needs_human verdicts; resumable on audited_file_sha256. Standalone trigger or chains…
Author a per-analyst-firm briefing deck (company overview + positioning + customer traction + product roadmap + competitive landscape + customer references + Q&A prep). CCO-Communications + CEO + CPO-Product co-authored. Use when user asks to "draft a analyst…
Author one or more Architecture Decision Records (adr@1) from an approved SRS plus architectural-spike@1 output (produced by architectural-spike-author; lean profile with no spike: the ADR options table carries the evidence inline). Use when user asks to…
Audit one or more existing automation-roadmap@1 markdowns against automation-roadmap_rubric@1.0. Produces a sibling .audit.md per artefact plus an AUDIT_BATCH_SUMMARY. Halts on needs_human verdicts; resumable on audited_file_sha256. Standalone trigger or…
Author the quarterly automation roadmap (opportunity portfolio + per-opportunity tech choice + ROI tracking + governance + risk register). UiPath / Automation Anywhere / Power Automate vendor patterns. Chief Automation Officer authored. Use when user asks to…
Out-of-band verification gate for the testing to done transition. Independently reruns the Task's section-1 cited tests plus its module suite via `awh eval` against a sealed, read-only baseline, and blocks the transition on any regression. Emits an awh-eval@1…
Audit one or more existing bias-audit@1 markdowns against bias-audit_rubric@1.0. Produces a sibling .audit.md per artefact plus an AUDIT_BATCH_SUMMARY. Halts on needs_human verdicts; resumable on audited_file_sha256. Standalone trigger or chains naturally…