aself101/agents-and-pipelines

aself101/agents-and-pipelines

Agent Claude Code

Identifies what a well-formed artifact of this type should contain but doesn't. Recognizes artifact types (API spec, agent definition, business plan, technical spec, policy document) and applies structural completeness criteria appropriate to that type. Distinguishes structural gaps (missing sections) from linkage…

aself101/agents-and-pipelines

Agent Claude Code

Performs Hegelian dialectical synthesis analysis on any artifact. Identifies dialectical pairs, detects false syntheses via the preservation test, traces developmental genealogies, and classifies current contradictions by developmental status. Decision - SYNTHESIZED/CONTRADICTORY.

aself101/agents-and-pipelines

Agent Claude Code

Performs Heraclitean unity-of-opposites analysis on any artifact. Identifies constitutive tensions, traces hidden flows, assesses logos coherence, and diagnoses stagnation. Decision - FLOWING/STAGNANT.

aself101/agents-and-pipelines

Agent Claude Code

Simulates someone whose interests are structurally opposed to the artifact succeeding. Generates the strongest available critique by modeling an interest-rival — not a security adversary, not a paradigm rival, but someone whose career advances when this artifact fails. Separates substance (what the hostile reader…

aself101/agents-and-pipelines

Agent Claude Code

Performs phenomenological reduction of a system's designed experience-model, followed by a fulfillment audit of the experience the system actually delivers. Brackets the design's experience-claims (epoché), audits every solicited act against delivered behavior (solicits -> intends -> delivers), maps projected…

aself101/agents-and-pipelines

Agent Claude Code

Reads the shape of what an artifact contains and infers what is structurally absent. Identifies holes the artifact's own architecture points to but doesn't fill. Produces a ranked absence inventory with structural evidence. Not feature requests but logic gaps. Decision - COMPLETE/IMPLIESMORE.

aself101/agents-and-pipelines

Agent Claude Code

Identifies what behaviors the artifact creates — intended or not. Every scoring framework, policy, workflow, and agent definition creates incentive gradients. Rational actors (human or automated) will optimize toward those gradients whether the author intended it or not. Makes the gradient visible before it's…

aself101/agents-and-pipelines

Agent Claude Code

Performs Machiavellian effectual truth analysis on any artifact. Reads the system's stated description and operational reality as two distinct sources, registers the gap between them. Inventories stated commitments, tests each against operational behavior, maps incentive flows and power structures, and audits…

aself101/agents-and-pipelines

Agent Claude Code

Models the perspective of a competent engineer inheriting the artifact five years downstream, with no original-author context. Identifies interpretive load — where the artifact assumes its reader shares tribal knowledge that was never written down. Surfaces why-not-documented friction - unexplained naming, orphaned…

aself101/agents-and-pipelines

Agent Claude Code

Reads meaning from what an artifact deliberately does not include. Every artifact has negative space — things the designer chose not to say, not to build, not to support. These silences are not gaps or oversights — they are design commitments expressed through absence. The negative space analyst identifies deliberate…

aself101/agents-and-pipelines

Agent Claude Code

Performs Nietzschean genealogical analysis on any artifact. Traces convention origins, maps power dynamics, classifies active vs reactive motivations, and assesses creative vitality. Decision - VITAL/DECADENT.

aself101/agents-and-pipelines

Agent Claude Code

Models the perspective of the person who runs the artifact in production — SREs, platform engineers, on-call rotation members. Reads through the lens of someone whose primary relationship is keeping it running, not building or using. Surfaces concerns neither builder nor user has vocabulary for - failure mode…

aself101/agents-and-pipelines

Agent Claude Code

Validates system performance, scalability, and resource efficiency under realistic load conditions. Tests response time profiling (P50, P95, P99), throughput capacity, concurrent user handling, resource usage, and degradation patterns. Executes actual load tests and measures real behavior under stress.

aself101/agents-and-pipelines

Agent Claude Code

Identifies failure modes that emerge when people or systems optimize against an artifact's measurable criteria — metric gaming, threshold satisficing, and ambiguity exploitation. Names the specific exploiting behaviors before they occur, so artifacts can be redesigned with eyes open. Decision - SOUND/PERVERSE.

aself101/agents-and-pipelines

Agent Claude Code

Reviews proposed designs BEFORE implementation begins. Validates architectural fit, design quality, scope appropriateness, and completeness. Catches design problems when they're cheap to fix. Provides PROCEED/REVISE decision. Blocks implementation if critical flaws exist or score < 80.

aself101/agents-and-pipelines

Agent Claude Code

Validates public-facing code quality including documentation completeness, feature coverage, unused code cleanup, and consumer experience. Ensures README reflects ALL shipped capabilities. Use AFTER code-validator passes. The "polish" gate for consumer experience.

aself101/agents-and-pipelines

Agent Claude Code

Final gate before publishing a package or CLI tool. Validates package.json, version consistency, documentation, exports, and release artifacts. Use AFTER all other validations pass, BEFORE npm publish or release.

aself101/agents-and-pipelines

Agent Claude Code

Validates actual runtime behavior of HTTP APIs, SDKs, and services through execution testing. Catches functional failures, incorrect responses, and integration issues. Starts with read-only GET endpoint verification, expandable to full behavioral testing.

aself101/agents-and-pipelines

Agent Claude Code

Comprehensive security auditor with risk assessment and numerical scoring. Use after implementation phases for pre-deployment security validation. Covers OWASP Top 10, CWE Top 25, and platform-specific vulnerabilities. Provides 1-100 score with explicit pass/fail thresholds.

aself101/agents-and-pipelines

Agent Claude Code

Performs Senecan failure trajectory projection on any artifact. Projects how unaddressed failure surfaces will evolve, which assumptions will be violated as conditions change, and where preparation gaps will widen. Produces a resilience trajectory with preparedness map. Decision - PREPARED/EXPOSED.

aself101/agents-and-pipelines

Agent Claude Code

Performs Socratic examination audit on any artifact. Diagnoses examination state through commitment extraction, contradiction classification, definition pressure testing, and confidence basis audit. Decision - EXAMINED/UNEXAMINED.

aself101/agents-and-pipelines

Agent Claude Code

Performs Socratic elenctic examination on any artifact. Extracts commitments, maps contradictions, pressure-tests definitions, and produces a structured inquiry agenda of questions the artifact cannot answer about itself. Decision - EXAMINED/UNEXAMINED.

aself101/agents-and-pipelines

Agent Claude Code

Performs software architecture evaluation on any artifact describing or embodying a software system. Assesses component boundaries, pattern fitness, specification-implementation correspondence, dependency structure, and architectural risk. Identifies where specifications mischaracterize the system they describe.…

aself101/agents-and-pipelines

Agent Claude Code

Validates stateful workflows and multi-step interactions work correctly across request sequences. Tests authentication flows, shopping carts, pagination, session management, and state transitions. Executes actual request sequences and verifies state persists and transitions correctly between steps.