ActiveMemory · josealekhine · May 11, 2026 · May 10, 2026 · May 10, 2026 · May 10, 2026
@@ -3,6 +3,7 @@
 <!-- INDEX:START -->
 | Date | Decision |
 |----|--------|
+| 2026-05-10 | Placeholder overrides use EXTEND not REPLACE semantics |
 | 2026-05-10 | Editorial constitution at .context/ingest/KB-RULES.md, not CONSTITUTION.md |
 | 2026-05-10 | Phase KB ships handover plus editorial paired, not split |
 | 2026-05-10 | KB ontology is pipeline-only-writer; no /ctx-kb-decide parallel skill |
@@ -134,6 +135,20 @@ For significant decisions:
 
 -->
 
+## [2026-05-10-181404] Placeholder overrides use EXTEND not REPLACE semantics
+
+**Status**: Accepted
+
+**Context**: When localizing the placeholder set used by validate.RejectPlaceholder, .ctxrc gains a placeholders: list. The existing precedent (rc.SessionPrefixes) uses REPLACE semantics: any non-empty user list completely replaces the shipped defaults. Placeholders need a different rule.
+
+**Decision**: Placeholder overrides use EXTEND not REPLACE semantics
+
+**Rationale**: The dominant case in this codebase is Tarzan Turkish — bilingual EN+TR projects where users need both English (TBD, n/a, see chat) and Turkish (iptal, yapılacak, görüşülecek) placeholders rejected simultaneously. REPLACE would force users to re-list every English default just to add one Turkish term, which they would skip and silently lose half the validator's coverage. EXTEND appends user list onto the shipped defaults so partial overrides do not regress baseline protection.
+
+**Consequence**: rc.Placeholders() must combine defaults + user list with case-folded de-duplication, diverging from the SessionPrefixes pattern. A future maintainer reading both accessors side-by-side will notice the inconsistency; the divergence is intentional and Spec: specs/placeholder-i18n.md captures why. If REPLACE is later wanted, add an opt-in placeholders_replace: true toggle rather than flipping the default.
+
+---
+
 ## [2026-05-10-001857] Editorial constitution at .context/ingest/KB-RULES.md, not CONSTITUTION.md
 
 **Status**: Accepted

@@ -17,6 +17,7 @@ DO NOT UPDATE FOR:
 <!-- INDEX:START -->
 | Date | Learning |
 |----|--------|
+| 2026-05-10 | Go compile/tool version mismatch comes from the cached toolchain, not the system Go |
 | 2026-05-10 | An ongoing user's concrete workaround tax is the strongest validation evidence |
 | 2026-05-10 | Lift renames alongside features when borrowing from battle-tested external designs |
 | 2026-05-10 | KB epistemology: in a KB you do not decide, you increase confidence |
@@ -129,6 +130,16 @@ DO NOT UPDATE FOR:
 
 ---
 
+## [2026-05-10-181418] Go compile/tool version mismatch comes from the cached toolchain, not the system Go
+
+**Context**: Hit 'compile: version "go1.26.1" does not match go tool version "go1.26.2"' on every go build / go test / make lint, even with my changes stashed out. System Go was 1.26.2 (healthy); go.mod pinned 1.26.1, so Go's auto-toolchain feature had downloaded 1.26.1 to ~/go/pkg/mod/golang.org/toolchain@v0.0.1-go1.26.1.darwin-arm64/. That cached toolchain was internally inconsistent: its compile binary and stdlib export data disagreed on version.
+
+**Lesson**: When the compile-vs-tool version error appears, the bug is the cached toolchain dir, not the installed Go. Reinstalling Go (brew, installer, etc.) does NOT touch the cached download, so the error persists after reinstall. Three real fixes: (1) rm -rf ~/go/pkg/mod/golang.org/toolchain@v0.0.1-go<X>.<platform>/ to force a clean re-download (~30s); (2) bump go.mod to match the system Go so the cached one is bypassed; (3) GOTOOLCHAIN=go<system version> to override the pin per-invocation. go clean -cache and GOTOOLCHAIN=local do not help.
+
+**Application**: First diagnostic on this error: check go env GOROOT — if it points to ~/go/pkg/mod/golang.org/toolchain@... the cached toolchain is in play. Then either delete the cached dir (most surgical) or bump go.mod (one-line diff, but lands in a commit). Do not waste time reinstalling Go.
+
+---
+
 ## [2026-05-10-001859] An ongoing user's concrete workaround tax is the strongest validation evidence
 
 **Context**: When extracting the editorial pipeline, the user pointed at things-wtf-disaster-recovery as a project where they were already running the editorial pattern manually — but at concrete cost: CLAUDE.md disabling half of ctx code-dev skills (/ctx-commit, /ctx-implement, /ctx-spec, /ctx-architecture, /ctx-brainstorm, /ctx-wrap-up), 10-CONSTITUTION.md at repo root colliding with .context/CONSTITUTION.md, hand-typed 8-item closeouts, hand-managed 20-INBOX.md, dedicated reference/vcf/external-grounding.md for ground-mode. The workaround was visible and the pain was specific.

@@ -32,6 +32,19 @@ TASK STATUS LABELS:
   `--phase` flag too, and we can have a auditor/normalizer for the current
   task document; or a skill that does a semantic pass, or both too.
 
+- [ ] Localize the placeholder set used by `RejectPlaceholder`
+  (decision add / learning add and any future body-flag validators).
+  Move the shipped defaults out of `internal/config/validate/placeholder.go`
+  Go constants into an embedded YAML asset, add a `.ctxrc placeholders:`
+  override with EXTEND semantics (user list is appended to defaults, not
+  replacing them — Tarzan Turkish is the dominant case), and replace the
+  current `strings.ToLower` with proper Unicode case folding via
+  `golang.org/x/text/cases` so İ/i, ß/SS, and similar fold correctly.
+  Ship `en` only in v1; ctx has no locale-specific assets yet, so the
+  structure is established but no `tr.yaml` lands in this work.
+  Spec: `specs/placeholder-i18n.md` #priority:high #added:2026-05-11
+  #prerequisite-for-locale-work
+
 ### Misc
 
 
@@ -1715,17 +1728,17 @@ Spec: `specs/ceremony-profiles.md`
 
 ### Phase SK: Skill Surface Polish (Phase 0a; prerequisite for Phase KB) `#priority:high #added:2026-05-09`
 
-Spec: TBD (design ref: `ideas/002-editorial-pipeline-and-skill-rigor.md` §3 "Reframing the wishy-washy skills")
+Spec: `specs/skill-surface-polish.md` (design ref: `ideas/002-editorial-pipeline-and-skill-rigor.md` §3 "Reframing the wishy-washy skills")
 
 Tightens existing capture skills to sibling-project rigor before the editorial pipeline (Phase KB) lifts that pattern wholesale. Independent of Phase RG; both can ship in parallel.
 
-- [ ] Add `MarkFlagRequired` to `ctx decision add` for `--context`, `--rationale`, `--consequence`; reject placeholder values (`TBD`, `see chat`, whitespace-only) at CLI level
-- [ ] Add `MarkFlagRequired` to `ctx learning add` for `--context`, `--lesson`, `--application`; same placeholder rejection
-- [ ] Add `--brief <path>` flag to `/ctx-spec` skill: when present, read the file as authoritative source per the sibling's authority order (frozen contracts > recorded decisions > debrief > agent inference labeled `TBD`); skip the fresh template Q&A
-- [ ] Update `/ctx-plan` skill to always offer to write the debated brief to `.context/briefs/<TS>-<slug>.md` at the end of an interview (creating `.context/briefs/` if absent)
-- [ ] Add an `Authority boundary (vs other skills)` section to `/ctx-decision-add`, `/ctx-learning-add`, `/ctx-task-add`, `/ctx-convention-add` skill files (prevent silent promotion handover→decision, learning→convention, etc., without explicit user ask)
-- [ ] Standardize "light compression for clarity is allowed; new facts are not" wording across capture skills (decide / learn primarily); same wording lands in `/ctx-handover` once Phase KB ships
-- [ ] Document the `--brief` contract in `docs/skills.md`
+- [x] Add `MarkFlagRequired` to `ctx decision add` for `--context`, `--rationale`, `--consequence`; reject placeholder values (`TBD`, `see chat`, whitespace-only) at CLI level
+- [x] Add `MarkFlagRequired` to `ctx learning add` for `--context`, `--lesson`, `--application`; same placeholder rejection
+- [x] Add `--brief <path>` flag to `/ctx-spec` skill: when present, read the file as authoritative source per the sibling's authority order (frozen contracts > recorded decisions > debrief > agent inference labeled `TBD`); skip the fresh template Q&A
+- [x] Update `/ctx-plan` skill to always offer to write the debated brief to `.context/briefs/<TS>-<slug>.md` at the end of an interview (creating `.context/briefs/` if absent)
+- [x] Add an `Authority boundary (vs other skills)` section to `/ctx-decision-add`, `/ctx-learning-add`, `/ctx-task-add`, `/ctx-convention-add` skill files (prevent silent promotion handover→decision, learning→convention, etc., without explicit user ask)
+- [x] Standardize "light compression for clarity is allowed; new facts are not" wording across capture skills (decide / learn primarily); same wording lands in `/ctx-handover` once Phase KB ships
+- [x] Document the `--brief` contract in `docs/skills.md` (landed in `docs/reference/skills.md` — the actual location)
 
 ### Phase RG: Require Git as Architectural Precondition (Phase 0b; prerequisite for Phase KB) `#priority:high #added:2026-05-09`
 

@@ -0,0 +1,190 @@
+---
+generated-at: 2026-05-11T00:08:47Z
+---
+# Handover [2026-05-11-000847] Phase SK shipped; 4 polish items on `internal/validate.BodyFlags` for next session
+
+**Provenance:** commit `971bf767` on branch `feat/skill-surface-polish`
+
+## Summary
+
+Phase SK (Skill Surface Polish) — all 7 tasks landed across
+6 commits on branch `feat/skill-surface-polish` (off `main` at
+`a44edfe3`):
+
+```
+971bf767 refactor(validate): consolidate body-flag helpers into internal/validate
+ba2faa54 refactor(validate): pure BodyFlags, no PreRunE decoration
+1156e44a docs(blog): align thought-piece bylines        (← unrelated, stacked intentionally)
+92507039 refactor(validate): single PreRunE enforcement, no panic, no swallowed errors
+55acbd81 feat(skills): /ctx-spec --brief, authority boundaries, plan brief offer
+f32c8fd9 feat(validate): require body flags on decision/learning add
+```
+
+All signed off, `make lint` 0 issues, `go test ./...` 0 failures,
+working tree clean. Branch is local-only (not pushed). Spec is at
+`specs/skill-surface-polish.md`; design ref at
+`ideas/002-editorial-pipeline-and-skill-rigor.md` §3.
+
+The 5-commit churn on the `validate` package (4 commits between
+`f32c8fd9` and `971bf767`) reflects iterative correction during
+the session — each commit removed a code-smell the previous one
+introduced. Functionally correct now, but the API surface still
+has 4 specific polish items the user surfaced at session end.
+They are non-blocking (the code works, the tests pass) but
+should be addressed before PR review.
+
+## Outstanding polish items (next session)
+
+All in `internal/validate/` and its two call sites
+(`internal/cli/decision/cmd/add/cmd.go`,
+`internal/cli/learning/cmd/add/cmd.go`):
+
+### 1. `RejectPlaceholder` should be unexported
+
+Currently exported as `validate.RejectPlaceholder`. Only call
+site is `BodyFlags` in the same package. Rename to
+`rejectPlaceholder`. Update test references.
+
+### 2. Per-file convention: private helper lives alone
+
+Convention check confirmed by `internal/sanitize/truncate.go`
+(single unexported `truncate` in its own file). After renaming,
+move the helper to `internal/validate/rejectplaceholder.go` and
+its tests to `internal/validate/rejectplaceholder_test.go`. The
+remaining `bodyflags.go` should hold only `BodyFlags`.
+
+### 3. `BodyFlags` takes too much
+
+```go
+// Current
+func BodyFlags(c *cobra.Command, flags ...string) error
+```
+
+Only uses `c.Flags()`. Change to:
+
+```go
+func BodyFlags(flags *pflag.FlagSet, names ...string) error
+```
+
+Call site:
+```go
+c.PreRunE = func(cobraCmd *cobra.Command, _ []string) error {
+    return validate.BodyFlags(cobraCmd.Flags(), cFlag.Context, ...)
+}
+```
+
+Update tests — the fixture no longer needs a full
+`cobra.Command`; constructing a `pflag.FlagSet` with two flags
+and calling `Parse(args)` is simpler.
+
+### 4. `Placeholders` map shape is confusing
+
+`internal/config/validate/placeholder.go` currently defines:
+
+```go
+const (
+    PlaceholderTBD = "tbd"
+    PlaceholderNA  = "n/a"
+    ...
+)
+
+var Placeholders = map[string]struct{}{
+    PlaceholderTBD: {},  // reads as if the key were the identifier
+    PlaceholderNA:  {},
+    ...
+}
+```
+
+In a map literal, `PlaceholderTBD:` evaluates to its const value
+`"tbd"` — the map key stored is the string, not the identifier.
+The user surfaced this as a real code smell: the surface reads
+"enum-keyed map" but it's actually a string-keyed set.
+
+**Resolution**: switch to a slice + linear scan (N=9; cost is
+negligible):
+
+```go
+var placeholders = []string{
+    PlaceholderTBD, PlaceholderNA, PlaceholderNAShort,
+    PlaceholderNone, PlaceholderSeeChat, PlaceholderSeeAbove,
+    PlaceholderSeeBelow, PlaceholderPending, PlaceholderToBeDone,
+}
+```
+
+```go
+// in rejectPlaceholder
+key := strings.ToLower(strings.TrimSpace(value))
+for _, p := range cfgValidate.Placeholders {
+    if key == p {
+        return errCli.FlagPlaceholder(flag, value)
+    }
+}
+```
+
+The slice's contents are the same constants, and the slice name
+documents the set's purpose without map-key sleight-of-hand. The
+audit's magic-strings check passes because every literal lives
+in `const PlaceholderXxx`.
+
+After the rename, `Placeholders` (capital P) should be
+`placeholders` (private) since only `internal/validate` reads it.
+
+## Gating
+
+- Each item, when fixed, requires the normal gate: `make lint`
+  clean, `go test ./...` clean, working tree clean before any
+  commit.
+- Sign every commit (`git commit -s`).
+- Do not add a `Co-Authored-By:` line to commits.
+- Stay on `feat/skill-surface-polish`; the branch is the
+  active feature branch and stacking polish on top is intended.
+- The 4 items are independent; one focused commit per item is
+  fine, or fold into a single `refactor(validate): polish surface`
+  commit if the diff stays small.
+
+## Session meta — read before resuming
+
+This session accumulated 5 saved feedback memories in
+`~/.claude/projects/-Users-volkan-Desktop-WORKSPACE-ctx/memory/`
+(2026-05-10), all from the same root cause: I acted on first
+impulse instead of grepping the codebase before scaffolding or
+editing. Specifically saved:
+
+- `feedback_no_coauthored_by.md` — strip Claude line only, never
+  the whole signoff block
+- `feedback_branch_before_commit.md` — branch off main first;
+  honour "stacking is intentional"
+- `feedback_always_signoff.md` — DCO requires `git commit -s`
+- `feedback_no_silent_errors_no_panic.md` — propagate errors, no
+  `_ =` or `panic` outside `Must`-prefixed functions
+- `feedback_no_silent_decoration.md` — helpers do not wrap
+  caller's cobra hooks
+- `feedback_grep_before_creating_packages.md` — extend existing
+  packages by default
+
+The user observed the cumulative drift mid-session and offered
+the handover. Resuming agent: **read those memory files first.**
+The 4 polish items above are technically small but every fix
+this session has introduced a new issue. Slow down: read the
+target file fully, grep for the convention, then edit.
+
+## Next session
+
+1. Pull this branch (`feat/skill-surface-polish @ 971bf767`).
+2. Read `internal/validate/bodyflags.go` and
+   `internal/config/validate/placeholder.go` end-to-end.
+3. Read `internal/sanitize/truncate.go` for the
+   single-private-helper-per-file pattern.
+4. Apply items 1–4 above. One or four commits, either is fine.
+5. After lint+test green and tree clean, hand back for push.
+
+Optional follow-up after the 4 fixes land: open the PR. PR body
+draft already prepared mid-session; see commit `55acbd81` body
+for the user-facing scope summary.
+
+## Open questions
+
+None. The 4 polish items have unambiguous resolutions above.
+The user-visible behaviour (decision/learning reject empty and
+placeholder body flags) is correct and tested. Polish is purely
+about code shape and API surface.
@@ -1,7 +1,7 @@
 ---
 title: "The Attention Budget: Why Your AI Forgets What You Just Told It"
 date: 2026-02-03
-author: Jose Alekhinne
+author: Volkan Özçelik
 reviewed_and_finalized: true
 topics:
   - attention mechanics
@@ -24,7 +24,7 @@ topics:
 
 ## Why Your AI Forgets What You Just Told It
 
-*Jose Alekhinne / 2026-02-03*
+*Volkan Özçelik / 2026-02-03*
 
 !!! question "Ever Wondered Why AI Gets Worse the Longer You Talk?"
     You paste a 2000-line file, explain the bug in detail, provide three

@@ -1,7 +1,7 @@
 ---
 title: "Skills That Fight the Platform"
 date: 2026-02-04
-author: Jose Alekhinne
+author: Volkan Özçelik
 reviewed_and_finalized: true
 topics:
   - context engineering
@@ -17,7 +17,7 @@ topics:
 
 ## When Your Custom Prompts Work against You
 
-*Jose Alekhinne / 2026-02-04*
+*Volkan Özçelik / 2026-02-04*
 
 !!! question "Have You Ever Written a Skill That Made Your AI Worse?"
     You craft detailed instructions. You add examples. You build elaborate

@@ -1,7 +1,7 @@
 ---
 title: "You Can't Import Expertise"
 date: 2026-02-05
-author: Jose Alekhinne
+author: Volkan Özçelik
 reviewed_and_finalized: true
 topics:
   - skill adaptation
@@ -17,7 +17,7 @@ topics:
 
 ## Why Good Skills Can't Be Copy-Pasted
 
-*Jose Alekhinne / 2026-02-05*
+*Volkan Özçelik / 2026-02-05*
 
 !!! question "Have You Ever Dropped a Well-Crafted Template into a Project and Had It Do... Nothing Useful?"
     * The template was **thorough**, 

@@ -1,7 +1,7 @@
 ---
 title: "Defense in Depth: Securing AI Agents"
 date: 2026-02-09
-author: Jose Alekhinne
+author: Volkan Özçelik
 reviewed_and_finalized: true
 topics:
   - agent security
@@ -17,7 +17,7 @@ topics:
 
 ## When Markdown Is Not a Security Boundary
 
-*Jose Alekhinne / 2026-02-09*
+*Volkan Özçelik / 2026-02-09*
 
 !!! question "What Happens When Your AI Agent Runs Overnight and Nobody Is Watching?"
     It follows instructions: **That is the problem**.

@@ -1,7 +1,7 @@
 ---
 title: "How Deep Is Too Deep?"
 date: 2026-02-12
-author: Jose Alekhinne
+author: Volkan Özçelik
 reviewed_and_finalized: true
 topics:
   - AI foundations
@@ -17,7 +17,7 @@ topics:
 
 ## When "Master ML" Is the Wrong Next Step
 
-*Jose Alekhinne / 2026-02-12*
+*Volkan Özçelik / 2026-02-12*
 
 !!! question "Have You Ever Felt like You Should Understand More of the Stack beneath You?"
     You can talk about transformers at a whiteboard.

@@ -1,7 +1,7 @@
 ---
 title: "Before Context Windows, We Had Bouncers"
 date: 2026-02-14
-author: Jose Alekhinne
+author: Volkan Özçelik
 reviewed_and_finalized: true
 topics:
   - context engineering