teleo/teleo-codex
8
0
Fork 0
Code Issues Pull requests 7 Projects Releases Packages Wiki Activity Actions 1

Add contributor docs, Alex brief, and evaluate-trigger #48

Merged
m3taversal merged 22 commits from leo/architecture-as-claims into main 2026-03-07 16:46:54 +00:00
Conversation 2 Commits 22 Files changed 52 +11979

22 commits

Author SHA1 Message Date
m3taversal
d1fa42bfc5 Fix agent naming: Theseus (not Logos) throughout 
Cory confirmed the AI alignment agent is Theseus. Reverted all
Logos references in skill file, evaluate-trigger, and Alex brief.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 16:42:40 +00:00
m3taversal
4be64979a0 Add contributor skill file and 2-agent evaluation trigger 
- .claude/skills/contribute/SKILL.md: installable skill for any Claude Code
  to contribute to Teleo Codex. Covers source ingestion, claim extraction,
  PR workflow, attribution, OPSEC rules.

- ops/evaluate-trigger.sh: upgraded to 2-agent review per PR:
  1. Leo (opus) — quality gates, cross-domain, coherence
  2. Domain agent (sonnet) — domain expertise, duplicates, technical accuracy
  Auto-detects domain from branch prefix or changed files.
  New flags: --leo-only, --dry-run shows detected agents.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 16:39:39 +00:00
m3taversal
bd9707a9cd Address Leo's review: 5 fixes to contributor docs 
1. Claim count: 160+ → 342+
2. Attribution: added "Your Credit" section to CONTRIBUTING.md
3. Theseus → Logos rename throughout ALEX_BRIEF.md
4. Contributor trailer for human proposers (Mode B)
5. OPSEC rules in both docs

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 16:32:08 +00:00
m3taversal
05ed5203f1 Add contributor docs, Alex onboarding brief, and evaluate-trigger script 
- CONTRIBUTING.md: step-by-step guide for external contributors
- docs/ALEX_BRIEF.md: onboarding brief for Alex — AI alignment domain
- ops/evaluate-trigger.sh: headless Leo evaluation trigger (Ganymede reviewed)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 16:22:25 +00:00
m3taversal
24fd456a09 Auto: 35 files | 35 files changed, 10533 insertions(+) 2026-03-07 15:10:14 +00:00
m3taversal
e36a46a361 leo: address Theseus + Rio review feedback on PR #45 
- Claim 3: expand from 146 auto-commits to 197/232 total (85% non-compliance) per Rio's audit — 50 manual commits without trailers are stronger evidence than auto-commits
- Claim 1: add implicit back-pressure on proposers (Rio's observation — bottleneck reshapes what work agents choose to do)
- Claim 2: add wiki link to evaluator bottleneck noting the interaction (Theseus — single evaluator + correlated priors compound)

Pentagon-Agent: Leo <76FB9BCA-CC16-4479-B3E5-25A3769B3D7E>

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 12:45:52 +00:00
m3taversal
e3e24b6e1b leo: 3 failure mode claims — evaluator bottleneck, correlated priors, social enforcement degradation 
- What: standalone claims documenting where the Teleo collective's architecture breaks today
- Why: PR #44's 10 operational claims painted only the success picture; Theseus flagged the absence of failure modes; these 3 are grounded in observed evidence (146 trailer-less auto-commits, single-evaluator review of all 44 PRs, zero cross-model reviews)
- Connections: complement the adversarial review, git trailer, and domain specialization claims by documenting their failure boundaries

Pentagon-Agent: Leo <76FB9BCA-CC16-4479-B3E5-25A3769B3D7E>

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 12:38:03 +00:00
m3taversal
f4852f35c3 Auto: core/living-agents/social enforcement of architectural rules degrades under tool pressure because automated systems that bypass conventions accumulate violations faster than review can catch them.md | 1 file changed, 58 insertions(+) 2026-03-07 12:37:47 +00:00
m3taversal
82476635b8 Auto: core/living-agents/all agents running the same model family creates correlated blind spots that adversarial review cannot catch because the evaluator shares the proposers training biases.md | 1 file changed, 64 insertions(+) 2026-03-07 12:37:05 +00:00
m3taversal
5f23712f70 Auto: core/living-agents/single evaluator bottleneck means review throughput scales linearly with proposer count because one agent reviewing every PR caps collective output at the evaluators context window.md | 1 file changed, 57 insertions(+) 2026-03-07 12:36:32 +00:00
m3taversal
f15d8a5ec5 leo: address review feedback from Rhea, Theseus, Rio on PR #44 
- Rhea: added structured author field to source archiving claim,
  fixed ghost email format to {id}@agents.livingip.ghost,
  added CI-as-enforcement as intermediate step before Forgejo ACLs
- Rio: fixed wiki link evidence (was not branch-timing, was nonexistent),
  corrected OPSEC timeline (rule came after files were written),
  fixed Doppler null-result (announcement article not whitepaper),
  removed duplicate Calypso/Vida reference

Pentagon-Agent: Leo <76FB9BCA-CC16-4479-B3E5-25A3769B3D7E>

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 12:06:45 +00:00
m3taversal
8a8a717825 leo: 10 architecture-as-claims — documenting how the Teleo collective works 
- What: 10 new claims in core/living-agents/ documenting the operational
  methodology of the Teleo collective as falsifiable claims, not instructions
- Why: The repo should document itself using its own format. Each claim
  grounds in evidence from 43 merged PRs, clearly separates what works
  today from what's planned, and identifies immediate improvements.
- Claims cover: PR review, prose-as-title, wiki-link graphs, domain
  specialization, confidence calibration, source archiving, git trailers,
  human-in-the-loop governance, musings, atomic notes
- This is Leo proposing about core/ — requires 2 domain agent reviews + Rhea

Pentagon-Agent: Leo <76FB9BCA-CC16-4479-B3E5-25A3769B3D7E>

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
 2026-03-07 12:01:13 +00:00
m3taversal
3b5cd0da90 Auto: core/living-agents/atomic notes with one claim per file enable independent evaluation and granular linking because bundled claims force reviewers to accept or reject unrelated propositions together.md | 1 file changed, 55 insertions(+) 2026-03-07 12:00:31 +00:00
m3taversal
a2eeacd06e Auto: core/living-agents/musings as pre-claim exploratory space let agents develop ideas without quality gate pressure because seeds that never mature are information not waste.md | 1 file changed, 52 insertions(+) 2026-03-07 12:00:04 +00:00
m3taversal
6a437a8ff1 Auto: core/living-agents/human-in-the-loop at the architectural level means humans set direction and approve structure while agents handle extraction synthesis and routine evaluation.md | 1 file changed, 67 insertions(+) 2026-03-07 11:59:33 +00:00
m3taversal
ead15d8bb9 Auto: core/living-agents/git trailers on a shared account solve multi-agent attribution because Pentagon-Agent headers in commit objects survive platform migration while GitHub-specific metadata does not.md | 1 file changed, 54 insertions(+) 2026-03-07 11:58:58 +00:00
m3taversal
6ef5bbb317 Auto: core/living-agents/source archiving with extraction provenance creates a complete audit trail from raw input to knowledge base output because every source records what was extracted and by whom.md | 1 file changed, 58 insertions(+) 2026-03-07 11:58:21 +00:00
m3taversal
ce7966ee99 Auto: core/living-agents/confidence calibration with four levels enforces honest uncertainty because proven requires strong evidence while speculative explicitly signals theoretical status.md | 1 file changed, 55 insertions(+) 2026-03-07 11:57:47 +00:00
m3taversal
6814a7c74d Auto: core/living-agents/domain specialization with cross-domain synthesis produces better collective intelligence than generalist agents because specialists build deeper knowledge while a dedicated synthesizer finds connections they cannot see from within their territory.md | 1 file changed, 63 insertions(+) 2026-03-07 11:57:16 +00:00
m3taversal
4de754580e Auto: core/living-agents/wiki-link graphs create auditable reasoning chains because every belief must cite claims and every position must cite beliefs making the path from evidence to conclusion traversable.md | 1 file changed, 56 insertions(+) 2026-03-07 11:56:41 +00:00
m3taversal
9654c2156d Auto: core/living-agents/prose-as-title forces claim specificity because a proposition that cannot be stated as a disagreeable sentence is not a real claim.md | 1 file changed, 61 insertions(+) 2026-03-07 11:56:12 +00:00
m3taversal
ce0dc81874 Auto: core/living-agents/adversarial PR review produces higher quality knowledge than self-review because separated proposer and evaluator roles catch errors that the originating agent cannot see.md | 1 file changed, 55 insertions(+) 2026-03-07 11:55:44 +00:00