Skip to main content
Jeden Skill in Manus ausführen
mit einem Klick
GitHub-Repository

crawfish

crawfish enthält 11 gesammelte Skills von crawfishai, mit Repository-Berufsabdeckung und Skill-Detailseiten auf SkillsMP.

gesammelte Skills
11
Stars
7
aktualisiert
2026-06-26
Forks
0
Berufsabdeckung
3 Berufskategorien · 100% klassifiziert
Repository-Explorer

Skills in diesem Repository

crawfish-authoring-definition-py
Softwareentwickler

Author definition.py — typed inputs/outputs (Parameter, static vs fluid) and the team shape (lead/coordination). Load when writing or editing a Definition's typed boundary. Static is the deliberate consequential choice; fluid is the untrusted default.

2026-06-26
crawfish-authoring-fixtures-evals
Softwarequalitätssicherungsanalysten und -tester

Author fixtures/ for craw test and gate on a baseline with craw eval — deterministic, mock-by-default, randomness carried in --seed. Load when adding test inputs or an eval gate to a Definition.

2026-06-26
crawfish-authoring-instructions-agents
Softwareentwickler

Author instructions.md and agents/*.md — front-matter (role, delegates_to, tools, model) over a markdown prompt. Load when writing a lead prompt or a subagent. Fluid inputs are data the prompt analyzes, never instructions to obey.

2026-06-26
crawfish-authoring-knowledge
Softwareentwickler

Attach knowledge with Wiki / with_context / SkillRef — summoned content is tainted data, never instructions, and is pinned by content hash. Load when giving an agent a rubric, corpus, or reference it should consult.

2026-06-26
crawfish-authoring-mcp-py
Softwareentwickler

Author mcp/*.py — MCPConnection instances; auth is a secret reference (env-var name), never inline. Load when adding an external tool server. A new connection re-enters the consent gate (craw code grant).

2026-06-26
crawfish-authoring-optimizing
Softwareentwickler

Optimize a Crawfish component — set a baseline (craw eval), search knobs (craw tune), iterate (craw refine), self-version (craw learn). Train vs eval mode, budget-bounded. Load when improving a component's quality or cost.

2026-06-26
crawfish-authoring-policies-skills
Softwareentwickler

Author policies/*.py (module-level Policy instances — consequential static config) and skills/*.md (bundled skills). Load when adding a guardrail/routing/permission policy or a bundled skill. A policy is static-only.

2026-06-26
crawfish-authoring-tools-py
Softwareentwickler

Author tools/*.py — a callable whose name matches the filename stem, with typed IO and taint propagation from fluid inputs. Load when writing a host-side tool. Agent-authored tool code compiles jailed (CRA-267), fail-closed.

2026-06-26
crawfish-determinism-ledger
Softwareentwickler

Determinism discipline and reading the ledger in Crawfish. Load before running, evaluating, or debugging a component: iterate on the mock (craw dev), carry randomness in --seed, promote to --live under --budget; read results via craw inspect / craw logs and the versioned --json surface.

2026-06-26
crawfish-pipeline-model
Softwareentwickler

The Crawfish pipeline mental model — Source → Filter → Batch → Aggregator → Router → Sink, and the Definition directory contract. Load when designing a pipeline or deciding between batch fan-out, an aggregator reduce, a router branch, or a refine loop.

2026-06-26
crawfish-security-spine
Informationssicherheitsanalysten

The Crawfish security spine — the prompt-injection boundary. Load whenever authoring or wiring a Definition, sink, MCP connection, policy, router, or pipeline. Fluid inputs are untrusted data and never reach a sink target or an instruction slot; consequential sink targets and idempotency keys are static-only; secrets resolve by reference, never in a prompt or config. This skill teaches the boundary; the craw code verbs enforce it.

2026-06-26