tkersey/agent-loop-schemes
codex/skills/agent-loop-schemes/SKILL.md
Translate an accepted GoalContract into normalized Zig actuation-operation topology. Use when work has repeated classes, migrations, debugging history, review campaigns, proof fanout, branch comparison, or nontrivial stopping conditions; remain advisory and never grant mutation or completion.
$ install --global
skillsauthnpx skillsauth add tkersey/dotfiles agent-loop-schemesInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
Security Scan Results
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
3
Scanners Passed
9
Scanners in report
Clean
TrivyContainer and dependency vulnerability scanner
95%
Clean
SemgrepStatic code analysis for vulnerabilities
95%
Clean
mcp-scan (Snyk)Model Context Protocol security validation
95%
Skipped
Snyk (dep)Open source security scanning
50%
Skipped
Socket.devSupply chain security analysis
50%
Skipped
VirusTotalMulti-engine malware detection
50%
Skipped
CrowdStrikeAdvanced threat intelligence
50%
Skipped
OSV-ScannerOpen Source Vulnerability database check
50%
Skipped
OWASP Dep-Check
50%
Last scanned: Jul 12, 2026, 4:28 AM290.2s2 files scanned
SKILL.md
- name:
- agent-loop-schemes
- description:
- Translate an accepted GoalContract into normalized Zig actuation-operation topology. Use when work has repeated classes, migrations, debugging history, review campaigns, proof fanout, branch comparison, or nontrivial stopping conditions; remain advisory and never grant mutation or completion.
Agent Loop Schemes
Mission
Describe the smallest step topology that $goal-actuating can place inside
actuation-open/v1 obligations and operations.
GoalContract
-> run-topology-advice/v1
-> $goal-actuating selects operations and obligation groups
This skill owns topology advice only. It does not create execution authority, run actions, fold evidence, count review attempts, or close the goal.
Advice shape
run_topology_advice:
version: run-topology-advice/v1
goal_id:
source_digest:
shape: direct | iterative | memoized-migration | debug-history | review-closeout | branch-comparison
step_producer:
rule:
owner_boundary:
decomposition:
step_template:
kinds: []
required_bindings: []
verifier_rule:
evidence_rule:
reducer: evidence-fold
continuation_requires_current_fold: true
review_rule:
classifier: review-fold
resolution: review-resolution/v1
raw_findings_are_steps: false
parallelism:
mode: none | scout-fanout | review-class-fanout | proof-fanout | branch-comparison
safe_when: []
fan_in_owner: goal-actuating
stop_rule:
success:
blocked:
invalid_proof:
invalidators: []
next_owner: goal-actuating
Selection
- Use
directfor one bounded step with one known verifier. - Use
iterativefor an ordinary dependency-ordered goal. - Use
memoized-migrationwhen many failures share one repair rule. - Use
debug-historywhen hypotheses or failure signatures may repeat. - Use
review-closeoutwhen findings must be classified, resolved, repaired, and reviewed to a current clean suffix. - Use
branch-comparisononly when isolated strategies share one verifier.
Laws
- Every material action becomes one selected run step.
- Every continuation consumes a current evidence fold.
- Review findings become resolution classes before they become work.
- Repeated owner-boundary pressure becomes one replacement-kernel node.
- Parallel leaves must be resource-disjoint and fold back through the lead.
- Public effects remain outside the topology and belong to
$ship. - Only a live closure decision completes the goal.
Procedure
- Bind advice to the accepted source digest and GoalContract.
- Choose one shape and one stop rule.
- Name the work producer, owner boundary, verifier, and evidence reducer.
- Preserve repeated-class memoization and invalid strategies.
- Describe only safe parallelism; shared owners remain serial.
- Hand advice to
$goal-actuating, which decides whether to project a WorkGraph.
Guardrails
- Do not invent a persistent controller.
- Do not emit executable work from raw review prose.
- Do not choose product scope or mutation authority.
- Do not create a second runtime beside
$goal-actuating. - Do not treat topology advice as proof or closure.
Related Skills
tkersey/fm
tools
VerifiedTrustedCommunity
Invokes Apple's macOS 27 fm command-line tool from a local Mac to use the on-device system model or Private Cloud Compute, including instructions, image prompts, schema-constrained JSON, and noninteractive automation. Use when the user asks to run Apple Foundation Models through fm, compare system versus pcc, generate structured output, or automate fm without Swift or an app.
64SKILL.mdUpdated Jul 20, 2026
tkersey/hylo
development
VerifiedTrustedCommunity
Compile historical Codex sessions into governed counterfactual evidence, evaluate an existing owner-applied candidate through blinded paired HCTP trials, and fold observable evidence into RUN, OBSERVE, or STOP. Use for `$hylo`, CRF extraction, counterfactual replay, source-governed direct or historical trials, sealed evidence, paired baseline/candidate evaluation, causal frontiers, or evidence-governed improvement.
64SKILL.mdUpdated Jul 13, 2026
tkersey/ledger
testing
VerifiedTrustedCommunity
Ensure a `ledger` command is available on PATH; materialize, validate, record, replay, and project requested Actuating artifacts without taking semantic or execution authority; coordinate the shared Learnings/Synesthesia/Negative Ledger lifecycle checkpoint and repo-local source-memory reconciliation; address Universalist plans and receipts; and perform pure artifact validation.
64SKILL.mdUpdated Jun 29, 2026
tkersey/review-fold
testing
VerifiedTrustedCommunity
Classify and quotient review findings, failing tests, incidents, bug reports, migration failures, and other witnessed falsifiers against accepted intent and the current Construction. Author counterexample-set/v1 without selecting repairs, counting review credit, or granting mutation.
64SKILL.mdUpdated Jun 28, 2026