self-improving-skills

Captures business administration issues, policy gaps, KPI misalignment, decision delays, handoff failures, and stakeholder misalignment to improve operational decision quality. Use when: (1) approval or execution bottlenecks appear, (2) KPI definitions conflict across teams, (3) process governance is inconsistent, (4) SLA commitments are missed or trending late, (5) budget variance requires triage, (6) vendor or cross-team handoff breaks, (7) policy documentation drifts from actual practice.

Captures smart-home automation conflicts, sensor drift, device connectivity failures, integration regressions, safety rule gaps, and energy optimization opportunities for continuous domotics improvement. Use when: (1) Automations conflict, loop, or misfire, (2) Sensors become stale or inaccurate, (3) Devices are unreachable or intermittently offline, (4) Cloud or local integrations break, (5) Occupancy detection is inconsistent with reality, (6) Latency causes delayed or jittery automations, (7) Energy usage patterns are inefficient, (8) Safety automations need stronger guardrails.

Improves the agent infrastructure itself — core prompt files (AGENTS.md, SOUL.md, TOOLS.md, CLAUDE.md), skill definitions, hooks, rules, extensions, and memory management. Use when: (1) An agent misinterprets a prompt file instruction, (2) A hook fails or doesn't trigger, (3) A skill is not activating correctly, (4) Rules conflict across files, (5) Context window is bloated by verbose prompt files, (6) Memory entries are stale or degrading quality, (7) A skill template is missing sections or unclear.

Captures negotiation strategy failures, concession leaks, BATNA weakness, framing misses, objection handling gaps, escalation misalignment, anchor errors, and agreement quality risks for continuous improvement. Use when negotiations stall, concessions exceed guardrails, terms are ambiguous, or recurring bargaining patterns emerge.

Captures forecast errors, supplier risks, logistics delays, inventory mismatches, quality deviations, and demand signal shifts to enable continuous supply chain improvement. Use when: (1) A stockout or backorder event occurs, (2) A delivery SLA is missed, (3) Supplier lead time increases, (4) Quality rejection rate spikes, (5) Demand forecast vs. actual variance exceeds 15%, (6) Warehouse capacity threshold is breached, (7) A procurement or routing decision needs documentation.

Captures learnings about GenAI/LLM configuration, model selection, inference optimization, fine-tuning, RAG pipelines, prompt engineering, multimodal processing, and cost management. Use when: (1) Model response quality degrades after a provider update or version change, (2) Inference latency exceeds acceptable thresholds, (3) Fine-tuned model regresses on evaluation benchmarks, (4) RAG retrieval returns irrelevant or stale chunks, (5) Token costs exceed budget projections, (6) Hallucination rate increases on factual queries, (7) Context window overflows cause critical information truncation, (8) Multimodal pipeline fails on specific input types (image, audio, video, PDF), (9) A better model or configuration is discovered for a task, (10) Guardrails block valid output or miss harmful content.

Captures data quality issues, metric drift, pipeline failures, misleading visualizations, metric definition mismatches, and data freshness problems to enable continuous analytics improvement. Use when: (1) An ETL/ELT pipeline fails, (2) A metric value shows anomalous behavior, (3) Two teams define the same metric differently, (4) A dashboard shows wrong or misleading data, (5) A data freshness SLA is missed, (6) A schema change breaks downstream consumers.

Captures lint errors, type mismatches, runtime bugs, anti-patterns, refactoring opportunities, language idiom gaps, debugging insights, and tooling issues to enable continuous coding improvement. Use when: (1) A lint or type error occurs, (2) A runtime exception is thrown, (3) An anti-pattern is identified in code, (4) A refactoring opportunity is discovered, (5) A better language idiom is found, (6) A debugging breakthrough reveals root cause, (7) A tooling issue blocks development.

Captures dialogue learnings, tone mismatches, escalation failures, and conversation quality issues for continuous improvement. Use when: (1) A user expresses frustration or confusion, (2) Tone mismatch is detected between agent and user, (3) Context is lost mid-conversation, (4) Agent hallucinates information, (5) User requests escalation to a human, (6) Conversation is abandoned or user rephrases repeatedly, (7) A missing conversational capability is identified. Also review learnings before handling complex dialogue flows.

Captures architecture decisions, code quality issues, build/deploy failures, dependency problems, performance regressions, tech debt accumulation, and test gaps for continuous engineering improvement. Use when: (1) A build or deployment fails, (2) An architecture violation is discovered, (3) A test gap or flaky test is found, (4) A dependency CVE or breaking change surfaces, (5) A performance regression is detected, (6) Code review reveals design flaws, (7) Tech debt accumulates past a threshold.

Captures reconciliation errors, forecast variances, control weaknesses, regulatory gaps, valuation errors, and cash flow anomalies to enable continuous finance operations improvement. Use when: (1) A reconciliation break is identified, (2) Budget vs. actual variance exceeds 10%, (3) A SOX control test fails, (4) A close task misses its deadline, (5) An intercompany imbalance is discovered, (6) An unusual journal entry is flagged by audit, (7) AR aging spikes past 90 days.

Captures policy gaps, compliance risks, recruiting process issues, onboarding friction, retention signals, candidate experience problems, and offboarding gaps to enable continuous HR improvement. Use when: (1) A compliance deadline is missed or approaching, (2) A candidate drops off during the recruiting pipeline, (3) A new hire leaves within 90 days, (4) A policy gap is discovered, (5) An exit interview reveals a recurring theme, (6) A benefits enrollment error occurs, (7) An I-9 or employment verification issue is found.

Captures clause risks, compliance gaps, precedent shifts, contract deviations, regulatory changes, and litigation exposure to enable continuous legal operations improvement. Use when: (1) An unfavorable clause is accepted or slips through review, (2) A compliance deadline is missed or approaching, (3) A new regulation impacts the organization, (4) A contract remains unsigned past SLA, (5) An IP infringement notice is received, (6) New case law changes interpretation in a relevant jurisdiction, (7) A data subject access request reveals process gaps.

Captures messaging misses, channel underperformance, audience drift, brand inconsistency, attribution gaps, and content decay to enable continuous marketing improvement. Use when: (1) CTR drops below threshold, (2) Conversion rate declines significantly, (3) Brand sentiment shifts negatively, (4) Organic traffic drops unexpectedly, (5) Email deliverability degrades, (6) UTM attribution breaks, (7) Campaign performance falls below benchmarks.

Captures process bottlenecks, incident patterns, capacity issues, automation gaps, SLA breaches, and toil accumulation to enable continuous operations improvement. Use when: (1) An incident repeats within 30 days, (2) MTTR exceeds target thresholds, (3) A manual step exists in an automated pipeline, (4) Alert fatigue indicates noisy monitoring, (5) Change failure rate spikes, (6) Toil exceeds 50% of on-call time.

Captures robotics autonomy failures, operational incidents, and engineering learnings to enable continuous improvement across perception, localization, planning, control, simulation, safety, and hardware integration. Use when: (1) Robot fails to localize in dynamic environment, (2) Planner fails in narrow passage or obstacle-rich scene, (3) Oscillatory control behavior or unstable PID tuning appears, (4) Sensor desync occurs (camera-lidar-imu timestamp mismatch), (5) Hardware driver drops packets or CAN timeout occurs, (6) Safety stop or emergency brake triggers unexpectedly, (7) Simulation succeeds but real robot fails, (8) Thermal throttling, battery sag, or power brownout appears.

Captures pipeline leaks, objection patterns, pricing errors, forecast misses, competitor shifts, and deal velocity drops to enable continuous sales improvement. Use when: (1) A deal slips past stage thresholds, (2) An objection recurs across multiple deals, (3) A pricing mistake is discovered, (4) Forecast accuracy falls below target, (5) A deal is lost to a competitor, (6) Discount escalations exceed policy, (7) A win/loss pattern emerges.

Captures learnings, experiment issues, and methodology corrections for continuous improvement in scientific research and ML workflows. Use when: (1) Data leakage detected in train/test split, (2) Model fails to reproduce across seeds or environments, (3) Statistical test misapplied or p-value misinterpreted, (4) Hypothesis test fails or needs revision, (5) Feature distribution shift detected, (6) User corrects methodology or analysis approach, (7) Experiment design flaw discovered. Also review learnings before designing new experiments.

Captures vulnerabilities, misconfigurations, access control violations, compliance gaps, incident response patterns, and threat intelligence to enable continuous security improvement. Use when: (1) A CVE or vulnerability is discovered, (2) Secrets are exposed in logs or output, (3) Access control violations or unauthorized access attempts occur, (4) Compliance audit findings or gaps are identified, (5) Security misconfigurations are found in infrastructure or applications, (6) Incident response procedures are executed or improved, (7) Threat intelligence is gathered from advisories or pen test results.

Captures ticket resolution delays, misdiagnoses, escalation gaps, SLA breaches, knowledge gaps, and customer churn signals to enable continuous support improvement. Use when: (1) A ticket is resolved late or incorrectly, (2) An SLA breach occurs, (3) A customer reopens a ticket, (4) An escalation pathway fails, (5) A knowledge base search returns no results, (6) CSAT scores drop below threshold, (7) A churn signal is detected in customer communication.