@lexwdex-org/opencode-dcp 3.2.0 → 3.2.2
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- package/README.md +1 -1
- package/dist/index.js +91 -111
- package/dist/index.js.map +1 -1
- package/dist/lib/prompts/compress-message.d.ts +1 -1
- package/dist/lib/prompts/compress-message.d.ts.map +1 -1
- package/dist/lib/prompts/compress-range.d.ts +1 -1
- package/dist/lib/prompts/compress-range.d.ts.map +1 -1
- package/dist/lib/prompts/system.d.ts +1 -1
- package/dist/lib/prompts/system.d.ts.map +1 -1
- package/package.json +1 -1
|
@@ -1,2 +1,2 @@
|
|
|
1
|
-
export declare const COMPRESS_MESSAGE = "
|
|
1
|
+
export declare const COMPRESS_MESSAGE = "Collapse selected individual messages in the conversation into detailed summaries.\n\nTHE SUMMARY\nYour summary must be EXHAUSTIVE. Capture file paths, function signatures, decisions made, constraints discovered, key findings, tool outcomes, and user intent details that matter... EVERYTHING that preserves the value of the selected message after the raw message is removed.\n\nUSER INTENT FIDELITY\nWhen a selected message contains user intent, preserve that intent with extra care. Do not change scope, constraints, priorities, acceptance criteria, or requested outcomes.\nDirectly quote short user instructions when that best preserves exact meaning.\n\nYet be LEAN. Strip away the noise: failed attempts that led nowhere, verbose tool output, and repetition. What remains should be pure signal - golden nuggets of detail that preserve full understanding with zero ambiguity.\nIf a message contains no significant technical decisions, code changes, or user requirements, produce a minimal one-line summary rather than a detailed one.\n\nMESSAGE IDS\nYou specify individual raw messages by ID using the injected IDs visible in the conversation:\n\n- `mNNNN` IDs identify raw messages\n\nEach message has an ID inside XML metadata tags like `<dcp-message-id priority=\"high\">m0007</dcp-message-id>`.\nThe same ID tag appears in every tool output of the message it belongs to \u2014 each unique ID identifies one complete message.\nTreat these tags as message metadata only, not as content to summarize. Use only the inner `mNNNN` value as the `messageId`.\nThe `priority` attribute indicates relative context cost. You MUST compress high-priority messages when their full text is no longer necessary for the active task.\nIf prior compress-tool results are present, always compress and summarize them minimally only as part of a broader compression pass. Do not invoke the compress tool solely to re-compress an earlier compression result.\nMessages marked as `<dcp-message-id>BLOCKED</dcp-message-id>` cannot be compressed.\n\nRules:\n\n- Pick each `messageId` directly from injected IDs visible in context.\n- Only use raw message IDs of the form `mNNNN`.\n- Ignore XML attributes such as `priority` when copying the ID; use only the inner `mNNNN` value.\n- Do not invent IDs. Use only IDs that are present in context.\n\nBATCHING\nSelect MANY messages in a single tool call when they are safe to compress.\nEach entry should summarize exactly one message, and the tool can receive as many entries as needed in one batch.\n\nGENERAL CLEANUP\nUse the topic \"general cleanup\" for broad cleanup passes.\nDuring general cleanup, compress all medium and high-priority messages that are not relevant to the active task.\nOptimize for reducing context footprint, not for grouping messages by topic.\nDo not compress away still-active instructions, unresolved questions, or constraints that are likely to matter soon.\nPrioritize the earliest messages in the context as they will be the least relevant to the active task.\nGeneral cleanup should be done periodically between other normal compression tool passes, not as the primary form of compression.\n";
|
|
2
2
|
//# sourceMappingURL=compress-message.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"compress-message.d.ts","sourceRoot":"","sources":["../../../lib/prompts/compress-message.ts"],"names":[],"mappings":"AAAA,eAAO,MAAM,gBAAgB,
|
|
1
|
+
{"version":3,"file":"compress-message.d.ts","sourceRoot":"","sources":["../../../lib/prompts/compress-message.ts"],"names":[],"mappings":"AAAA,eAAO,MAAM,gBAAgB,qlGA0C5B,CAAA"}
|
|
@@ -1,2 +1,2 @@
|
|
|
1
|
-
export declare const COMPRESS_RANGE = "
|
|
1
|
+
export declare const COMPRESS_RANGE = "Collapse a range in the conversation into a detailed summary.\n\nTHE SUMMARY\nYour summary must be EXHAUSTIVE. Capture file paths, function signatures, decisions made, constraints discovered, key findings... EVERYTHING that maintains context integrity. This is not a brief note - it is an authoritative record so faithful that the original conversation adds no value.\n\nUSER INTENT FIDELITY\nWhen the compressed range includes user messages, preserve the user's intent with extra care. Do not change scope, constraints, priorities, acceptance criteria, or requested outcomes.\nDirectly quote user messages when they are short enough to include safely. Direct quotes are preferred when they best preserve exact meaning.\n\nYet be LEAN. Strip away the noise: failed attempts that led nowhere, verbose tool outputs, back-and-forth exploration. What remains should be pure signal - golden nuggets of detail that preserve full understanding with zero ambiguity.\n\nCOMPRESSED BLOCK PLACEHOLDERS\nWhen the selected range includes previously compressed blocks, use this exact placeholder format when referencing one:\n\n- `(bN)`\n\nCompressed block sections in context are clearly marked with a header:\n\n- `[Compressed conversation section]`\n\nCompressed block IDs always use the `bN` form (never `mNNNN`) and are represented in the same XML metadata tag format.\n\nRules:\n\n- Include every required block placeholder exactly once.\n- Do not invent placeholders for blocks outside the selected range.\n- Treat `(bN)` placeholders as RESERVED TOKENS. Do not emit `(bN)` text anywhere except intentional placeholders.\n- If you need to mention a block in prose, use plain text like `compressed bN` (not as a placeholder).\n- Preflight check before finalizing: the set of `(bN)` placeholders in your summary must exactly match the required set, with no duplicates.\n\nThese placeholders are semantic references. They will be replaced with the full stored compressed block content when the tool processes your output.\n\nFLOW PRESERVATION WITH PLACEHOLDERS\nWhen you use compressed block placeholders, write the surrounding summary text so it still reads correctly AFTER placeholder expansion.\n\n- Treat each placeholder as a stand-in for a full conversation segment, not as a short label.\n- Ensure transitions before and after each placeholder preserve chronology and causality.\n- Do not write text that depends on the placeholder staying literal (for example, \"as noted in `(b2)`\").\n- Your final meaning must be coherent once each placeholder is replaced with its full compressed block content.\n\nBOUNDARY IDS\nYou specify boundaries by ID using the injected IDs visible in the conversation:\n\n- `mNNNN` IDs identify raw messages\n- `bN` IDs identify previously compressed blocks\n\nEach message has an ID inside XML metadata tags like `<dcp-message-id>...</dcp-message-id>`.\nThe same ID tag appears in every tool output of the message it belongs to \u2014 each unique ID identifies one complete message.\nTreat these tags as boundary metadata only, not as tool result content.\n\nRules:\n\n- Pick `startId` and `endId` directly from injected IDs in context.\n- IDs must exist in the current visible context.\n- `startId` must appear before `endId`.\n- Do not invent IDs. Use only IDs that are present in context.\n\nBATCHING\nWhen multiple independent ranges are ready and their boundaries do not overlap, include all of them as separate entries in the `content` array of a single tool call. Each entry should have its own `startId`, `endId`, and `summary`.\n";
|
|
2
2
|
//# sourceMappingURL=compress-range.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"compress-range.d.ts","sourceRoot":"","sources":["../../../lib/prompts/compress-range.ts"],"names":[],"mappings":"AAAA,eAAO,MAAM,cAAc,
|
|
1
|
+
{"version":3,"file":"compress-range.d.ts","sourceRoot":"","sources":["../../../lib/prompts/compress-range.ts"],"names":[],"mappings":"AAAA,eAAO,MAAM,cAAc,k/GA2D1B,CAAA"}
|
|
@@ -1,2 +1,2 @@
|
|
|
1
|
-
export declare const SYSTEM = "\
|
|
1
|
+
export declare const SYSTEM = "\nYou operate in a context-constrained environment. Manage context continuously to avoid buildup and preserve retrieval quality. Efficient context management is paramount for your agentic performance.\n\nThe ONLY tool you have for context management is `compress`. It replaces older conversation content with technical summaries you produce.\n\n`<dcp-message-id>` and `<dcp-system-reminder>` tags are environment-injected metadata. Do not output them.\n\nTHE PHILOSOPHY OF COMPRESS\n`compress` transforms conversation content into dense, high-fidelity summaries. This is not cleanup - it is crystallization. Your summary becomes the authoritative record of what transpired.\n\nThink of compression as phase transitions: raw exploration becomes refined understanding. The original context served its purpose; your summary now carries that understanding forward.\n\nCOMPRESS WHEN\n\nA section is genuinely closed and the raw conversation has served its purpose:\n\n- Research concluded and findings are clear\n- Implementation finished and verified\n- Exploration exhausted and patterns understood\n- Dead-end noise can be discarded without waiting for a whole chapter to close\n\nDO NOT COMPRESS IF\n\n- Raw context is still relevant and needed for edits or precise references\n- The target content is still actively in progress\n- You may need exact code, error messages, or file contents in the immediate next steps\n\nBefore compressing, ask: _\"Is this section closed enough to become summary-only right now?\"_\n\nEvaluate conversation signal-to-noise REGULARLY. Use `compress` deliberately with quality-first summaries. Prioritize stale content intelligently to maintain a high-signal context window that supports your agency.\n\nIt is of your responsibility to keep a sharp, high-quality context window for optimal performance.\n";
|
|
2
2
|
//# sourceMappingURL=system.d.ts.map
|
|
@@ -1 +1 @@
|
|
|
1
|
-
{"version":3,"file":"system.d.ts","sourceRoot":"","sources":["../../../lib/prompts/system.ts"],"names":[],"mappings":"AAAA,eAAO,MAAM,MAAM,
|
|
1
|
+
{"version":3,"file":"system.d.ts","sourceRoot":"","sources":["../../../lib/prompts/system.ts"],"names":[],"mappings":"AAAA,eAAO,MAAM,MAAM,gzDAgClB,CAAA"}
|
package/package.json
CHANGED
|
@@ -1,7 +1,7 @@
|
|
|
1
1
|
{
|
|
2
2
|
"$schema": "https://json.schemastore.org/package.json",
|
|
3
3
|
"name": "@lexwdex-org/opencode-dcp",
|
|
4
|
-
"version": "3.2.
|
|
4
|
+
"version": "3.2.2",
|
|
5
5
|
"type": "module",
|
|
6
6
|
"description": "OpenCode plugin that optimizes token usage by pruning obsolete tool outputs from conversation context",
|
|
7
7
|
"main": "./dist/index.js",
|