cocotte1992 · Oct 13, 2026 · Oct 13, 2026 · Oct 13, 2026 · Oct 13, 2026 · Oct 13, 2026
diff --git a/.github/copilot-instructions.md b/.github/copilot-instructions.md
@@ -52,6 +52,7 @@ Failure shapes to catch yourself in, and the command that replaces each:
 - not knowing which command answers a question you can already state in words → answer "<question>", which resolves the subject in the index, runs the right command, prints via: <command>, and refuses rather than guess
 - a shell text search with context flags to find a function body → read "file::symbol"
 - paging one function with view/view_range → read "file::symbol"
+- issuing sequential or overlapping line-range reads across a file (paging) → read "file::symbol" / outline file / skeleton file / rg <pattern> <file>
 - a read refused by a [tg] deny, retried with a shifted range, different lines, or another read tool → never retry reading the file at different lines; run the command the deny names or use bash-output / token-goat read
 - reading a symbol plus chasing its callers and containing doc section as separate reads → brief "file::symbol"
 - reading one heading of a large doc → section "file::Heading"

diff --git a/CHANGELOG.md b/CHANGELOG.md
diff --git a/CLAUDE.arch.md b/CLAUDE.arch.md
diff --git a/README.md b/README.md
diff --git a/docs/architecture-qa.html b/docs/architecture-qa.html
@@ -1283,7 +1283,7 @@ <h1>Token-Goat Architecture & Engineering Q&A</h1>
 </p>
 <div style="font-size:12px; color:var(--text-dim); line-height:1.5;">
 <strong>Step 1: Verify Command Compression &amp; Fidelity (Automated Suite):</strong><br>
-Run <code>node dist/token-goat.mjs bench</code> in your terminal. This replays a corpus of captured command output through the compressors and prints the byte-savings ratio plus a fidelity check (it exits 1 if a must-keep line is dropped). At the time of writing this ran at 96.1% saved with 6/6 fidelity checks kept; the ratio is not a fixed guarantee, only a floor the corpus must clear on every run.<br><br>
+Run <code>node dist/token-goat.mjs bench</code> in your terminal. This replays a corpus of captured command output through the compressors and prints the byte-savings ratio plus a fidelity check (it exits 1 if a must-keep line is dropped). At the time of writing this ran at 87.2% saved, counting the fence and recall line a compressed result carries, with 6/6 fidelity checks kept; the ratio is not a fixed guarantee, only a floor the corpus must clear on every run.<br><br>
 
 <strong>Step 2: Verify Polyglot Surgical-Read Savings (Unit Benchmark):</strong><br>
 Run <code>npx vitest run <a href="https://github.com/DFKHelper/token-goat/blob/main/tests/token_savings_benchmark.test.ts" target="_blank" class="github-inline-ref">tests/token_savings_benchmark.test.ts</a></code>. This measures real TypeScript, Python, Go, and Markdown fixtures and enforces a floor (60% average, with a documented per-case exception at 50% for Go's doc-comment-preserving outline) below the 85-97% range <code>docs/architecture.md</code> quotes; the measured average at the time of writing was 80.2%, and the test prints the actual per-case numbers rather than asserting a fixed figure.<br><br>

diff --git a/docs/cli.md b/docs/cli.md
diff --git a/docs/hook-channel-matrix.md b/docs/hook-channel-matrix.md
diff --git a/docs/install.md b/docs/install.md
@@ -148,7 +148,7 @@ The hooks were checked loading and running under agy 1.2.11 on Windows, and the
 token-goat install --opencode
 ```
 
-The `--opencode` flag drops a TypeScript bridge plugin into opencode's plugins directory. Post-edit indexing, compact assist, and rewritten tool results (prompt-injection fencing, secret redaction, and output compression replace the raw result, the same protection Claude Code sessions get) work. So do repeat-search denial for `websearch`, repeat-load denial for `skill`, and the subagent prompt briefing for `task` — all three tool ids and their argument keys were verified against opencode's own source at the installed release's tag. Session hints don't — opencode's plugin API has no way to inject context before a tool read. Neither does image shrinking: the only way to hand opencode a smaller image is to point its `read` at a copy, and opencode checks its `read` and outside-directory permissions against that copy's path instead of the image's, so a rule you set on the image would stop applying. Shell commands are not wrapped for compression either, since opencode's permission rules can come from an organization account or a well-known URL no hook reads; their output is still compressed after they run.
+The `--opencode` flag drops a TypeScript bridge plugin into opencode's plugins directory. Post-edit indexing, compact assist, and rewritten tool results (prompt-injection fencing, secret redaction, and output compression replace the raw result, the same protection Claude Code sessions get) work, MCP tool results included, except an MCP result the server marks as an error, which opencode raises as an error before any plugin hook sees it. So do repeat-search denial for `websearch`, repeat-load denial for `skill`, and the subagent prompt briefing for `task` — all three tool ids and their argument keys were verified against opencode's own source at the installed release's tag. Session hints don't — opencode's plugin API has no way to inject context before a tool read. Neither does image shrinking: the only way to hand opencode a smaller image is to point its `read` at a copy, and opencode checks its `read` and outside-directory permissions against that copy's path instead of the image's, so a rule you set on the image would stop applying. Shell commands are not wrapped for compression either, since opencode's permission rules can come from an organization account or a well-known URL no hook reads; their output is still compressed after they run.
 
 ### openclaw users
 

diff --git a/docs/security.md b/docs/security.md
diff --git a/eslint-suppressions.json b/eslint-suppressions.json
@@ -1,7 +1,7 @@
 {
   "src/bash_extractors.ts": {
     "regexp/no-super-linear-backtracking": {
-      "count": 19
+      "count": 17
     }
   },
   "src/cli_bootstrap_audit.ts": {

diff --git a/package-lock.json b/package-lock.json
diff --git a/package.json b/package.json
@@ -143,7 +143,6 @@
     "unzipper": "^0.12.5",
     "adm-zip": "^0.6.1",
     "fast-uri": "^3.1.6",
-    "qs": "^6.16.0",
-    "tesseract.js-core": "6.1.2"
+    "qs": "^6.16.0"
   }
 }