diff --git a/CONTRIBUTING.md b/CONTRIBUTING.md index fa060dd..ce0fae1 100644 --- a/CONTRIBUTING.md +++ b/CONTRIBUTING.md @@ -2,7 +2,7 @@ # Contributing -Boatstack is a generated content distribution. Propose changes to workflow semantics, templates, evidence rules, or generated presentation in [Intelligence Flow](https://github.com/operatorstack/intelligence-flow/tree/3c770b5db125ad5f28515380c59c4d1718e3a020/labs/12-product-engineering-loop). +Boatstack is a generated content distribution. Propose changes to workflow semantics, templates, evidence rules, or generated presentation in [Intelligence Flow](https://github.com/operatorstack/intelligence-flow/tree/9a3cafb4f5040ed787390607b7cd6ced843588ea/labs/12-product-engineering-loop). The Boatstack repository receives product/runtime changes through a generated pull request. Review the PR's `UPSTREAM.json`, tests, adapter diff, and context-size change; do not hand-edit generated output on `main`. `.github/workflows` is the exception: it is Boatstack's executable control plane, excluded from scheduled projection and changed only through a separate manually reviewed Boatstack PR. diff --git a/UPSTREAM.json b/UPSTREAM.json index 952cdaa..b103bc7 100644 --- a/UPSTREAM.json +++ b/UPSTREAM.json @@ -12,7 +12,7 @@ }, "files": { ".gitignore": "a7079e923a776f14f1bb3a6aa0a11a133a8e1dfb35af020f327623357b7e3957", - "CONTRIBUTING.md": "57832885d79ec4dc777699fd52028e297bf787041a927299b38cbf3d51bb5a21", + "CONTRIBUTING.md": "c0ea4700fd74f0930812adde111a8719f13697963d027905f71a6251a68a548f", "README.md": "3ce3e95e511089b44e946a44b8d5f4f81d019ece5336db65b2cab1f9dc4d4dad", "assets/boatstack-journey.svg": "e465befc50c8ce30f3e07e8fd97012931beeb053392c8fbf38ad645023b3cc63", "assets/boatstack-mark.svg": "be1f984da1bfa69fa5d1f986d8343d21f7e20921b71db888c928b4d2e54b09b5", @@ -160,8 +160,8 @@ "boatstack/runtime_cache.go": "e026ffc1906f7e1e98b768bae63e6658164d2826c07169c9121ce0f23c73faf8", "boatstack/runtime_cache_test.go": "b981467ddc9f0f562da6bff5de7a80a9fe5a433a0317541d1e48df268546ac85", "boatstack/runtime_provenance_test.go": "1d52f1e6b0691cf4667729cc9b9f3c55c128f0aa3321f3a2843a9aa6fd0e73dc", - "boatstack/safety.go": "31dbeb58ff22359812fee8e4113d47ebb563733908d17eddb264fe9e92fe2ca7", - "boatstack/safety_test.go": "c5c5413e66266a5159e56e6f9681fc5fdce4c4e53174f207c1605b639bd9d686", + "boatstack/safety.go": "e21cc46048e51ed973096a4c4f1c32908aff79cfba8f2f75a301c5f6521656c8", + "boatstack/safety_test.go": "500ad53cd5e3a700553eb781d9eaf4028ae27478796759a76bfd57321fff5a7c", "boatstack/safety_update_publisher_test.go": "ed3f8187036623694dfe7c395cdae00fdae14609bab6124d1fdfc6fe73fa2196", "boatstack/skill_frontmatter.go": "73364df463ce828c2d005aab55f72bb92f7a34d99cf3f53d4e0cd5a4da9dbd0e", "boatstack/skill_frontmatter_test.go": "a3ec52e7df357a72265c95dd66db15d9c0effc7e5f90f14ce69c27792ce394eb", @@ -187,10 +187,10 @@ "docs/benchmark-corpus-audit.md": "f2d206fe8579a514f9da82b2c96c19b343ac004be67617e1bd34f0f8e0e5e6c6", "docs/benchmark-submission-audit.md": "9518abdd17690729c6423f87cab20418ed47b0915b5faa44b9ef975e9e9c3b79", "docs/configuration.md": "060775c73431f28bd16066bdf9e0f89034d2855c7ca0f5544f660d24b91211d0", - "docs/evidence-engineered-coding.md": "0c0e344f5eae535c5b406ed1bff51c807fd31aa7768668491888d3f9f84df293", + "docs/evidence-engineered-coding.md": "1e76e625f5e891abdc08384ef3fc91559232c9e8a43f2c64a61798990f122337", "docs/generated-files.md": "437791765b0a4015032ae21d1a6618563cad92b7402819e4f963bf5ae16284a3", "docs/getting-started.md": "51c2823f21e35140d31e6d5083dc4b89fddd24721ac6acc474154a4da53ee9f8", - "docs/public-claims.json": "1cf62f1ffa0d5fadd49c7c4843143a09ffa38a1d78309de590378542a5a8ea78", + "docs/public-claims.json": "cfa12bb1b55a8d1a7a5e70565298caaae19151e972d31bb16820128dac92cb7f", "docs/public-surface.md": "713f7a050b5f339cf948299103ef3800417dccfecf2cc1a4166397ea6f978907", "docs/research-and-design.md": "8d78678108f0a6c924e1ff9b32c0f81aae9d1f779e0082843b6f99ad993ae2b6", "docs/safety.md": "7b9b5c515d36e683767ec8d3d9d6d119ac93650b2f629d351deadd4c600ed6a6", @@ -204,7 +204,7 @@ "labs/diagram-json/compiled/evidence.md": "1ba1c989ade070a8ef9a508fbd788d100d7292f2dbacbb2bce895468019f619d", "labs/diagram-json/compiled/tasks.json": "88f60851abf79d851e9fccc754ff3040034ae595306bc87d64784c19eb403e71", "labs/diagram-json/compiled/test-matrix.json": "424657ff505768e50fa113801fd8363364a18269d5297480907a993d44063a39", - "labs/diagram-json/plan.lock.json": "dc95b0ece06634bae69a6f010a0d1fb49c6b2c43659031a29b78c891466deade", + "labs/diagram-json/plan.lock.json": "96fd7a684a7324c8443881e4defc991e25846f83f41eac6a027f3129337fb188", "labs/diagram-json/plan.md": "3cc4f533b8d69386deff16b3a594a3ba09d4c0c3db636cccd8c4380084ce6a51", "labs/diagram-json/questions.md": "74733b015002c8a6777c558e7e997fa48c94850b9bd39054fe9366c97ecf728d", "labs/diagram-json/request.md": "0808fc41c36779c404f4a3a121167da6e76cac56df526e70f9ed6d3e0d4c02ed", @@ -318,6 +318,7 @@ "release-notes/2026-07-26-guard-hydrate-double-check.md": "83a5591aba6bf30c9f4008ba8d26bf1994ef3fd61145f46fcdd1678912b9990b", "release-notes/2026-07-26-hidden-jflow-design-note.md": "f60ed9dbbfb46a172ac9d33dd758a3166f820007b1673029f29f0fbefa0e5c0a", "release-notes/2026-07-26-workspace-reap.md": "e691d6a1c232cf218157880655413005fcb2c4f3113ededffdb80899a5054bb8", + "release-notes/2026-07-27-constitutional-boundary-floor.md": "41514cbea53867c264e354f149638444ab0d80f64896b9cf720f3979ae78b3de", "release-notes/2026-07-27-coreachable-recovery.md": "6ffc6b0e9a7d46c0f99a64112813c33d19571c73d02e98ac5573924f1663fd54", "release-notes/2026-07-27-discoverable-planning-errors.md": "d8099d1a6cd1805c3fcd446d9fa95739dec93e57ff421ecf4f99562b143c9966", "release-notes/2026-07-27-invalid-delivery-block-actionable.md": "8fac8e3921e2285291703efa46e624b72cb5bac1b8492beca4c4b633abb5ba16", @@ -326,7 +327,7 @@ "generator": "operatorstack/intelligence-flow:boatstack-distribution", "schema_version": 1, "source": { - "commit": "3c770b5db125ad5f28515380c59c4d1718e3a020", + "commit": "9a3cafb4f5040ed787390607b7cd6ced843588ea", "path": "labs/12-product-engineering-loop", "repository": "operatorstack/intelligence-flow" } diff --git a/boatstack/safety.go b/boatstack/safety.go index 55ada0e..ea7a59f 100644 --- a/boatstack/safety.go +++ b/boatstack/safety.go @@ -60,6 +60,16 @@ func malformedHookInput(code string) error { // still banned in isPureReadOnlyCommand, so no filter can be turned into a writer. var readOnlyStage = regexp.MustCompile(`(?i)^\s*(?:env\s+[^ ]+\s+)*(?:rg|grep|git\s+(?:grep|diff|status|show|log)|cat|sed|head|tail|less|wc|awk|sort|uniq|cut|tr|jq|column|nl|comm|rev|fold|find\s+[^\n]*-(?:print|ls)|psql\s+[^\n]*\s-c\s+["']?\s*select\b|(?:[^\s]*/)?boatstack-helper(?:[_.-][a-z0-9._-]+)?\s+(?:recovery-status|mutation-status|operation-status|delivery-status|next-status|workspace-status|repair-status|check-plan|check-source-plan|check-safety|diagnose-hook|doctor|version)\b)`) +// Constitutional/Optimization split. These destruction rules are CONSTITUTIONAL: +// they define the real boundary (destroying a live resource) and are never traded +// for convenience — no project config knob disables them; config may only ADD +// scope (HighRiskPaths). The executor-gating around them (the scanSQL argument to +// classifySafetyText; the executed-vs-data file distinction; the artifact exemption +// on committed diffs) is the OPTIMIZATION surface: it narrows WHEN a rule is +// observed to cut false positives, but it may never disable the boundary — when the +// executor is live, a constitutional rule must still fire. That floor is enforced by +// TestExecutorGatingNeverDisablesTheBoundary. +// // irreversiblePatterns classify destruction by text. Rules whose regex names its // own EXECUTOR (rm, git, terraform, supabase db reset, …) are self-executing: // matching the text is sound because the text IS the command. Rules marked diff --git a/boatstack/safety_test.go b/boatstack/safety_test.go index eb3b833..e921645 100644 --- a/boatstack/safety_test.go +++ b/boatstack/safety_test.go @@ -198,6 +198,43 @@ func TestReadOnlyInspectionPipelinesAllowed(t *testing.T) { } } +// Constitutional/Optimization split: the executor-gating optimization narrows WHEN +// a destruction rule is observed (to cut false positives) but must NEVER disable the +// boundary. When the executor is live the constitutional rule still fires; the +// optimization's benefit (data operations pass) does not become a leak; and no +// project config can trade the boundary away. +func TestExecutorGatingNeverDisablesTheBoundary(t *testing.T) { + repo := safetyTestRepo(t) + if err := os.WriteFile(filepath.Join(repo, "migrate.sql"), []byte("DROP TABLE accounts;\n"), 0o644); err != nil { + t.Fatal(err) + } + // Optimization FLOOR: a live executor still blocks (the gating did not disable it). + mustBlock := []string{ + `psql -c "DROP TABLE accounts"`, // executor-gated SQL, live client + `psql -f migrate.sql`, // live client runs a DDL file it executes + `rm -rf /`, // self-executing constitutional + `terraform destroy -auto-approve`, // self-executing constitutional + `supabase db reset`, // self-executing constitutional + `git push --force origin main`, // self-executing constitutional + } + for _, command := range mustBlock { + if findings := ClassifyCommand(repo, command); len(findings) == 0 { + t.Errorf("optimization disabled the boundary — live destruction allowed: %q", command) + } + } + // Optimization BENEFIT (not a leak): the same DDL as inert data passes, because + // no live executor observes it. + mustPass := []string{ + "git add migrate.sql", + `git commit -m "add migration with DROP TABLE accounts"`, + } + for _, command := range mustPass { + if findings := ClassifyCommand(repo, command); len(findings) != 0 { + t.Errorf("data operation wrongly blocked: %q -> %#v", command, findings) + } + } +} + func TestSafeDiagnosticsAndFixForwardCommandsRemainAllowed(t *testing.T) { repo := safetyTestRepo(t) safeScript := filepath.Join(repo, "scripts", "apply_schema.py") diff --git a/docs/evidence-engineered-coding.md b/docs/evidence-engineered-coding.md index 3c5b9cd..a3c58fd 100644 --- a/docs/evidence-engineered-coding.md +++ b/docs/evidence-engineered-coding.md @@ -146,6 +146,6 @@ Delivery and system improvement also remain separate. A failed task may suggest ## What is evidence-backed -The current moves were derived from the Intelligence Flow benchmark corpus and product-repository studies. The generated source commit is [`3c770b5db125ad5f28515380c59c4d1718e3a020`](https://github.com/operatorstack/intelligence-flow/tree/3c770b5db125ad5f28515380c59c4d1718e3a020/labs/12-product-engineering-loop). +The current moves were derived from the Intelligence Flow benchmark corpus and product-repository studies. The generated source commit is [`9a3cafb4f5040ed787390607b7cd6ced843588ea`](https://github.com/operatorstack/intelligence-flow/tree/9a3cafb4f5040ed787390607b7cd6ced843588ea/labs/12-product-engineering-loop). The evidence supports specific failure mechanisms and guardrails. It does not establish that Boatstack is optimal, that control-theory notation proves software quality, or that one workflow dominates every team. Those are evaluation questions, so the distribution preserves measurements, provenance, gaps, and negative results. diff --git a/docs/public-claims.json b/docs/public-claims.json index c0da935..d742fd3 100644 --- a/docs/public-claims.json +++ b/docs/public-claims.json @@ -1,6 +1,6 @@ { "schema_version": 1, - "source_commit": "3c770b5db125ad5f28515380c59c4d1718e3a020", + "source_commit": "9a3cafb4f5040ed787390607b7cd6ced843588ea", "statuses": ["verified", "observed", "still_being_evaluated"], "claims": [ { @@ -12,7 +12,7 @@ "readable_evidence": "why-these-steps.md#portable-workflow-and-state", "implementation": ["../boatstack/export.go", "../boatstack/references/artifacts.md", "../boatstack/references/workflow.md"], "verification": ["../boatstack/export_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "human-decisions", @@ -23,7 +23,7 @@ "readable_evidence": "why-these-steps.md#human-decisions", "implementation": ["../boatstack/references/workflow.md", "../boatstack/plan.go"], "verification": ["../boatstack/plan_test.go", "../boatstack/planning_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "validation-provenance", @@ -34,7 +34,7 @@ "readable_evidence": "why-these-steps.md#validation-provenance", "implementation": ["validation-and-evidence.md", "../boatstack/plan.go"], "verification": ["../boatstack/plan_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "irreversible-operations", @@ -46,7 +46,7 @@ "readable_evidence": "why-these-steps.md#irreversible-operations", "implementation": ["safety.md", "../boatstack/safety.go", "../boatstack/hooks.go"], "verification": ["../boatstack/safety_test.go", "../boatstack/hooks_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "reviewer-ready-pr", @@ -57,7 +57,7 @@ "readable_evidence": "why-these-steps.md#reviewer-ready-pr", "implementation": ["../boatstack/pr.go", "getting-started.md"], "verification": ["../boatstack/pr_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "phase-scoped-delivery", @@ -68,7 +68,7 @@ "readable_evidence": "why-these-steps.md#phase-scoped-delivery", "implementation": ["../boatstack/delivery.go", "../boatstack/safety.go", "../boatstack/hooks.go", "../boatstack/references/workflow.md"], "verification": ["../boatstack/delivery_test.go", "../boatstack/pr_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "model-neutral-contract", @@ -79,7 +79,7 @@ "readable_evidence": "why-these-steps.md#model-choice-and-budget", "implementation": ["research-and-design.md", "../boatstack/references/workflow.md"], "verification": ["../boatstack/export_test.go", "../boatstack/planning_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "cross-model-failures", @@ -90,7 +90,7 @@ "readable_evidence": "why-these-steps.md#model-choice-and-budget", "implementation": ["research-and-design.md"], "verification": ["benchmark-corpus-audit.md", "benchmark-submission-audit.md"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "lower-cost-outcomes", @@ -101,7 +101,7 @@ "readable_evidence": "why-these-steps.md#model-choice-and-budget", "implementation": ["research-and-design.md"], "verification": ["benchmark-corpus-audit.md", "benchmark-submission-audit.md"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "git-worktree-activation", @@ -112,7 +112,7 @@ "readable_evidence": "why-these-steps.md#git-worktree-activation", "implementation": ["../boatstack/runtime_cache.go", "../boatstack/hooks.go"], "verification": ["../boatstack/runtime_cache_test.go", "../boatstack/hooks_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" }, { "id": "visible-updates", @@ -123,7 +123,7 @@ "readable_evidence": "why-these-steps.md#visible-updates", "implementation": ["../boatstack/update.go", "../boatstack/init.go"], "verification": ["../boatstack/update_test.go", "../boatstack/init_test.go", "../boatstack/export_test.go"], - "last_verified_version": "source:3c770b5db125ad5f28515380c59c4d1718e3a020" + "last_verified_version": "source:9a3cafb4f5040ed787390607b7cd6ced843588ea" } ] } diff --git a/labs/diagram-json/plan.lock.json b/labs/diagram-json/plan.lock.json index c812981..d5f4591 100644 --- a/labs/diagram-json/plan.lock.json +++ b/labs/diagram-json/plan.lock.json @@ -6,7 +6,7 @@ "plan_path": "labs/diagram-json/plan.md", "plan_sha256": "3cc4f533b8d69386deff16b3a594a3ba09d4c0c3db636cccd8c4380084ce6a51", "schema_version": 1, - "source_commit": "3c770b5db125ad5f28515380c59c4d1718e3a020", + "source_commit": "9a3cafb4f5040ed787390607b7cd6ced843588ea", "source_plan_path": "labs/diagram-json/source-plan.md", "source_plan_sha256": "e10593ddaa7522ab80cc991d0a09399257139799e37f737794cd49d68a39985b", "spec_path": "labs/diagram-json/spec.md", diff --git a/release-notes/2026-07-27-constitutional-boundary-floor.md b/release-notes/2026-07-27-constitutional-boundary-floor.md new file mode 100644 index 0000000..ae05868 --- /dev/null +++ b/release-notes/2026-07-27-constitutional-boundary-floor.md @@ -0,0 +1,16 @@ +### The destruction boundary is guaranteed to survive its own optimization + +The guard now decides destruction by effect: it asks which executor runs a command, not whether a file +or a message spells a keyword. That change removed false positives — committing a migration, staging a +schema file, editing a note. This release records and enforces the limit of that optimization. + +The rule that decides destruction is constitutional. It defines the real boundary — destroying a live +resource — and is never traded for convenience. No project setting disables it; a setting can only add +scope. The executor check that narrows when the rule is observed is an optimization: it exists to cut +false positives, and it may never let a live destructive command through. When the executor is live, the +rule still fires. + +A conformance test now holds this floor. A live database client running a drop, a client running a +migration file, a recursive delete, an infrastructure destroy, a database reset, and a force push are +all still denied, while the same SQL sitting inert in a staged or committed file passes. The benefit of +the optimization does not become a leak, and the boundary cannot be optimized away.