docs(prompts): codex master prompts — PR #53 rebase + workflows Infisical integration #72
No reviewers
Labels
No labels
W6d-automerge-calibration
agent/claude-code
agent/codex
agent/hermes
agent/iskra
agent/ollama
agent/patchwarden
automerge-candidate
class/security-sensitive
cutover-gate
dependency/blocked
dependency/blocks-others
dependency/cross-repo
dependency/needs-confirmation
domain:agents
domain:ci
domain:docs
domain:forgejo
domain:infra
domain:memory
domain:runtime
domain:signal
domain:ux
flow/architecture
flow/blocked
flow/deployed
flow/done
flow/implementation
flow/intake
flow/maintained
flow/observed
flow/ready
flow/refining
flow/retired
flow/review
iterating
judge/codex-candidate
judge/hermes-candidate
judge/low-confidence
judge/needs-refinement
judge/operator-needed
judge/p0
judge/p1
judge/p2
judge/p3
judge/park
judge/patchwarden-candidate
judge/stale-priority
kind/adr
kind/bug
kind/chore
kind/feature
kind/infra
kind/ops
kind/refactor
kind/research
large-impact
merge/auto
merge/manual
merge/manual-dependency-conflict
merge/manual-failing-tests
merge/manual-merge-conflict
merge/manual-missing-review
merge/manual-operator-preference
merge/manual-red-zone
merge/manual-security-sensitive
merge/manual-unclear-scope
merge/manual-unknown
meta
mode:operator-only
mode:patchwarden-iskra-approved
mode:safe-auto
needs-operator-decision
needs-triage
not-ready
observed/erroring
observed/needs-followup
observed/pending
observed/retire-candidate
observed/unused
observed/used
operator-emotional
owner-attention
phase/02
phase/03
priority:p0
priority:p1
priority:p2
priority:p3
proposed
ready-for-agent
ready-for-operator
recovery
review:claude-reviewed
review:codex-reviewed
review:dziadek-reviewed
review:needs-human
risk/exposure
risk/process
risk/product
risk/runtime
safety:external-write
safety:no-prod-mutation
safety:prod-impact
safety:secret-touch
size/large
size/medium
size/small
size/tiny
size/unknown
source/adr
source/agent-generated
source/manual
source/operator-chat
source/voice-note
status:blocked
status:codex-ready
status:merged:pending-evidence
status:needs-evidence
status:operator-needed
status:parked
tier/full
tier/lite
tier/stacked
tier:0-platform-substrate
tier:1-iskra-value-layer
tier:2-tools-products-modules
type:bug
type:chore
type:docs
type:feat
type:policy
type:research
No milestone
No project
No assignees
3 participants
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
pdurlej/platform!72
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "claude/orders/codex-prompts-canary-ops"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Canary status: missing — Medium PR class (governance + Codex master prompts driving CI changes); fire canary 3+3 manually before merge
Purpose
Two master prompts for Codex GPT-5.5 execution, operationalizing canary workflow per operator decisions 2026-05-05:
prompts/codex-pr-53-rebase.md— small operational fix. Codex rebases PR #53 on current main, resolves.forgejo/workflows/canary-required.ymlconflict per #53 design, force-pushes withcodexidentity. Unblocks PR #53 merge.prompts/workflows-infisical-integration.md— feature work. Codex replaces Forgejo repo secrets in canary workflow with Infisical service-token retrieval. Operator chose this over manual sync (option b over option a) because "Infisical jako single source of truth" matches charter §5 direction.This PR does NOT do
.forgejo/workflows/canary-required.ymldirectly (Codex changes it)Changed artifacts
prompts/codex-pr-53-rebase.md(new, ~150 lines) — rebase + conflict resolution policy, identity check, escape hatches, self-verificationprompts/workflows-infisical-integration.md(new, ~210 lines) — pre-work research (R1 inventory, R2 Infisical structure verify, R3 CLI vs SDK decision, R4 bootstrap-secret scope), implementation steps S1-S5, self-verification, PR opening conventions, escape hatchesTotal: 2 files, +362 lines.
Why now
Relevant context
PLATFORM_CHARTER.md§5 — Infisical primary, Vault sunset; this work fulfills the principle in workflow domainmigrations/vault-to-infisical.md— Vault sunset migration plan; Infisical structure should already be set up per Phase 0AGENTS.md§"Identity-isolation" + §"Joining as a new agent" — both prompts reference these for identity disciplinestate/agent-execution-template.md(just landed in PR #69) — referenced by both prompts for execution protocolRuntime evidence
N/A — pure docs/prompts change. No workflow modification in this PR. Codex's resulting PRs (one per prompt) will have runtime evidence.
Known constraints
iskra-platform, envprod, path/forgejo-actions/canary). Codex's R2 step verifies these against actual Infisical config; if assumptions wrong, Codex stops and surfaces.platform-wt-takeover-pr51still exists with codex/orders/canary-workflow-rewrite branch; if not, Codex needs to recreate from origin.Explicit out-of-scope
Requested decision
approve_mergeafter canary 3+3 fires (manual; workflow this PR depends on for auto-fire is not yet operational).Merge blockers
How operator triggers Codex
After this PR merges, in master operator (Codex thread):
Orchestrator (claude) reports back when both PRs land.
Spec sources read
70cde4579) — to verify #53 architecture works (collect-diff GREEN + canary honest-red)PLATFORM_CHARTER.md§5 — operator's Infisical primary stancedecisions/0002-ci-enforcement-canary.md— for trust boundary requirements that prompt 2 must preserveAGENTS.md§"Identity-isolation" + §"Joining as a new agent" — referenced by both promptsstate/agent-execution-template.md— referenced for execution protocolTest plan
ZAI_API_KEY+CANARY_FORGEJO_TOKENat predictable path (or names where they live)codex exec < prompts/codex-pr-53-rebase.mdfirst; PR #53 conflict resolvedcodex exec < prompts/workflows-infisical-integration.md; new PR opens with Infisical integrationcodex(NOT pdurlej; verified viauser.login)INFISICAL_SERVICE_TOKENto Forgejo repo secrets (Issue #49 amended setup task)Superseded by #89 and the post-#89 operating model.
The prompts here were useful as planning artifacts, but they now encode stale instructions around #53, force-push/rebase flow, and direct workflow secret assumptions. Merging them as agent-facing docs would make future agents more likely to choose the wrong path.
The live follow-up is a new implementation plan/PR for a runner-local Infisical bridge plus canary evidence, not these historical prompts. Closing as superseded.
Pull request closed