Skip to content
Merged
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension


Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
21 changes: 0 additions & 21 deletions .github/workflows/learning-loop.yml
Original file line number Diff line number Diff line change
Expand Up @@ -40,14 +40,6 @@ jobs:
echo "🧠 Generating rule proposals from observed patterns..."
bash fleet-coordinator.sh generate-rules

- name: Upload rule proposals
if: always()
uses: actions/upload-artifact@v7.0.2
with:
name: rule-proposals
path: shared-context/learning/rule-proposals/*.lgt
if-no-files-found: ignore

- name: Show learning summary
run: |
echo "## Learning Loop Summary" >> $GITHUB_STEP_SUMMARY
Expand All @@ -59,19 +51,6 @@ jobs:
echo "- No observations yet" >> $GITHUB_STEP_SUMMARY
fi

if [ -d shared-context/learning/rule-proposals ]; then
PROPOSALS=$(find shared-context/learning/rule-proposals -name "*.lgt" | wc -l)
echo "- Rule proposals: $PROPOSALS" >> $GITHUB_STEP_SUMMARY

if [ $PROPOSALS -gt 0 ]; then
echo "" >> $GITHUB_STEP_SUMMARY
echo "### Proposed Rules" >> $GITHUB_STEP_SUMMARY
for rule in shared-context/learning/rule-proposals/*.lgt; do
[ -f "$rule" ] && echo "- $(basename "$rule")" >> $GITHUB_STEP_SUMMARY
done
fi
fi

deploy-approved-rules:
name: Deploy Approved Rules to Hypatia
runs-on: ubuntu-latest
Expand Down
1 change: 0 additions & 1 deletion .gitignore
Original file line number Diff line number Diff line change
Expand Up @@ -154,7 +154,6 @@ pnpm-lock.yaml
.editorconfig
.env.*
/exports/*.json
/exports/*.lgt
/logs/
ai-cli-crash-capture/
composer/*.beam
Expand Down
1 change: 0 additions & 1 deletion .machine_readable/descriptiles/NEUROSYM.a2ml
Original file line number Diff line number Diff line change
Expand Up @@ -11,7 +11,6 @@ last-updated = "2026-04-11"
[hypatia-config]
scan-enabled = true
scan-depth = "standard" # quick | standard | deep
report-format = "logtalk"

[symbolic-rules]
# Custom symbolic rules for this project
Expand Down
4 changes: 2 additions & 2 deletions audits/directive-census-2026-09-26/REPORT.md
Original file line number Diff line number Diff line change
Expand Up @@ -64,9 +64,9 @@ Remote snapshots are pinned individually, not a simultaneous estate-wide snapsho

### 3. Hypatia's local instructions conflict with the fleet's old learning output

[Hypatia AGENTIC](https://github.com/hyperpolymath/hypatia/blob/9f2f62f5c9463c79b33a5ebf54372166ce56f349/.machine_readable/descriptiles/AGENTIC.a2ml) explicitly prohibits Logtalk, says rules are pure Elixir, and prohibits emitting JSON from its tooling. It requires evidence, reruns and behavioral tests, and declares automation rate/confidence constraints.
[Hypatia AGENTIC](https://github.com/hyperpolymath/hypatia/blob/9f2f62f5c9463c79b33a5ebf54372166ce56f349/.machine_readable/descriptiles/AGENTIC.a2ml) explicitly prohibits the retired rule language, says rules are pure Elixir, and prohibits emitting JSON from its tooling. It requires evidence, reruns and behavioral tests, and declares automation rate/confidence constraints.

The audited fleet coordinator still generates Logtalk `.lgt` proposals and has a publication path to Hypatia. Its local quarantine now blocks that route; the census did not deploy that block.
The audited fleet coordinator still generated rule-file proposals (the generator has since been removed) and had a publication path to Hypatia. Its local quarantine now blocks that route; the census did not deploy that block.

The same AGENTIC file references retired `6a2` locations while acknowledging the newer `descriptiles` convention. **Do not automatically resolve this by deleting the local constraints or regenerating the file.** Separate the obsolete location reference from still-binding substantive prohibitions.

Expand Down
2 changes: 1 addition & 1 deletion audits/fleet-wiring-2026-09-26.md
Original file line number Diff line number Diff line change
Expand Up @@ -89,7 +89,7 @@ No centralized repository-policy reconciliation across all these entry points wa

`tests/e2e.sh` describes and primarily checks repository structure, shell syntax and required configuration. A passing “E2E Tests” run is not proof of dispatch → bot execution → changed target → verification → published check/PR.

The learning workflow explicitly disables `deploy-approved-rules`; the coordinator's generic generated Logtalk detector contains a TODO and an unconditional `fail`. Generated proposals are not necessarily functioning new detectors.
The learning workflow explicitly disables `deploy-approved-rules`; the coordinator's generic generated detector contained a TODO and an unconditional `fail`. Generated proposals are not necessarily functioning new detectors.

## Live GitHub observations and limits

Expand Down
2 changes: 1 addition & 1 deletion bots/finishingbot/ROADMAP.adoc
Original file line number Diff line number Diff line change
Expand Up @@ -91,7 +91,7 @@ This roadmap outlines the development plan for finishing-bot, the release readin
**Target:** 2026-Q3

=== Hypatia Integration
* Logtalk rule-based validation
* Rule-based validation (Hypatia rules)
* Formal verification of release criteria
* Proof-carrying code validation
* Rule synthesis from validation history
Expand Down
148 changes: 0 additions & 148 deletions bots/glambot/bot-integration/src/glambot_findings.lgt

This file was deleted.

10 changes: 5 additions & 5 deletions docs/BOT-OPERATIONS.adoc
Original file line number Diff line number Diff line change
Expand Up @@ -72,7 +72,7 @@ coordination layer:
|`+process-findings+` |Process `+findings/<repo>/latest.json+`, dispatch
auto-fixes, trigger learning

|`+generate-rules+` |Analyze pattern frequency, propose Logtalk rules at
|`+generate-rules+` |Analyze pattern frequency, (retired) rule proposals at
threshold (5+ observations)

|`+deploy-bots+` |Generate deployment status JSON
Expand All @@ -84,7 +84,7 @@ threshold (5+ observations)

....
Findings → observed-patterns.jsonl → threshold (5+) → propose_new_rule()
→ Logtalk rule in learning/rule-proposals/ → create_rule_approval_pr()
→ (retired: rule proposal and approval PR)
→ Human reviews PR → Approved rule deployed to hypatia
....

Expand Down Expand Up @@ -366,15 +366,15 @@ license/placeholder - Integration tests
|*Purpose* |Neurosymbolic CI/CD intelligence platform
|*Repo* |`+hyperpolymath/hypatia+`
|*Path* |`+$REPOS_DIR/hypatia/+`
|*Tech* |Rust + Haskell + Logtalk/Prolog + Ada/SPARK
|*Tech* |Rust + Haskell + Ada/SPARK
|*Tier* |0 (Engine)
|*Completion* |*70%*
|*Dependencies* |None (coordinates everything)
|===

*Components*: - *CLI* (Rust) - 7 commands: scan, deposit, withdraw,
search, batch, fleet, hooks - *Registry* (Haskell) - Type-safe ruleset
DSL, property-based testing - *Engine* (Logtalk) - Declarative rule
DSL, property-based testing - *Engine* - Declarative rule
execution, distillation - *Adapters* (Rust) - GitHub, GitLab, Bitbucket,
Codeberg (SourceHut/Radicle pending) - *Data* (Rust) - ArangoDB graph
DB, Dragonfly cache - *Fixer* (Rust) - Error catalog parser, issue
Expand Down Expand Up @@ -461,7 +461,7 @@ git-dispatcher (Central Coordination Hub)
│ └── shared-context (Rust coordination library)
│
├── hypatia (Rules Engine, Engine tier)
│ ├── Logtalk rules (symbolic reasoning)
│ ├── Elixir rules (symbolic reasoning)
│ ├── Haskell registry (type-safe verification)
│ └── ArangoDB + Dragonfly (graph DB + cache)
│
Expand Down
8 changes: 1 addition & 7 deletions docs/archive/DOGFOODING-ANALYSIS.adoc
Original file line number Diff line number Diff line change
Expand Up @@ -52,7 +52,7 @@ public for transparency 3. Publish findings after patches deployed
|===
|Repo |Language |Unsafe Patterns Found |Severity |Can Self-Fix?
|*hypatia* |Rust |27 unwrap calls |Medium |✅ YES
|*gitbot-fleet* |Shell/Logtalk |3 getExn, 3 Obj.magic |Low |✅ YES
|*gitbot-fleet* |Shell |3 getExn, 3 Obj.magic |Low |✅ YES
|*rhodibot* |Rust |0 issues |None |✅ CLEAN
|*echidnabot* |Rust |6 unwrap calls |Medium |✅ YES
|*glambot* |Rust |1 unwrap call |Low |✅ YES
Expand Down Expand Up @@ -88,9 +88,6 @@ review before merging
* [ ] *Add hypatia self-scan to CI/CD*
** Workflow: `+.github/workflows/hypatia-self-scan.yml+`
** Block merges if critical issues found
* [ ] *Document all Logtalk predicates*
** Many predicates in `+engine/scanner.lgt+` lack comments
** Add usage examples for each rule in `+code-safety-lessons.lgt+`
* [ ] *Test suite for learning engine*
** Test observation thresholds (5 for proposal, 10 for auto-approve)
** Test fix outcome tracking
Expand All @@ -101,9 +98,6 @@ review before merging
* [ ] *Create proper Rust library crate*
** Current structure mixes CLI, adapters, fixer, scanner
** Separate: `+hypatia-core+`, `+hypatia-cli+`, `+hypatia-adapters+`
* [ ] *Add SWI-Prolog installation to CI*
** Currently using bash POC scanner
** Full Logtalk validation requires SWI-Prolog
* [ ] *Implement rule versioning*
** Track which hypatia version generated each rule
** Support rule deprecation and migration
Expand Down
2 changes: 1 addition & 1 deletion docs/archive/NEW-PATTERNS-DEPLOYMENT-2026-02-06.adoc
Original file line number Diff line number Diff line change
Expand Up @@ -137,7 +137,7 @@ reached for new patterns
==== Immediate (Ready Now)

[arabic]
. ✅ Generate Logtalk rule proposals for 3 new patterns
. ✅ Generate rule proposals for 3 new patterns
. ✅ Submit for ECHIDNA validation
. ✅ Human review and approval

Expand Down
Loading
Loading